Unigine::Image Class

Header:

#include <UnigineImage.h>

List of functions:

Interface for handling textures and other images.

Textures are stored on CPU.

The color of each pixel of an image is represented by the Pixel structure. Pixel format is determined by the image file format (see the table below).

Image type	Pixel format
8-bit unsigned formats: R8, RG8, RGB8, RGBA8	Pixel color components are in range from 0 to 255. E.g. Pixel(10, 10, 255, 255) for RGBA8 Color components are accessed via *Pixel.i*
16-bit unsigned formats: R16, RG16, RGB16, RGBA16	Pixel color components are in range from 0 to 65535. Color components are accessed via *Pixel.i*
16-bit float formats: R16F, RG16F, RGB16F, RGBA16F	Pixel color components are in range from 0.0f to 1.0f. Color components are accessed via *Pixel.i*
32-bit float formats: R32F, RG32F, RGB32F, RGBA32F	Pixel color components are in range from 0.0f to 1.0f. Color components are accessed via *Pixel.f*

Usage example

This example demostrates how to:

Create an image.
Set color information for each pixel of the image using the Pixel structure.
Save the image to a file.
Get current image format.
Convert the image to another format.
Get color information for each pixel of the image using the Pixel structure.
Clear the image.

Source code (C++)

// AppWorldLogic.cpp

#include <UnigineMaterials.h>

/* .. */

int AppWorldLogic::init() {
	// image size parameters
	const int ASPECT_RATIO = 1;
	const int width = 256;
	const int height = width / ASPECT_RATIO;

	ImagePtr image = Image::create();

	// creating an image of the specified size
	image->create2D(width, height, Image::FORMAT_RGBA8);

	// declaring a pixel to store color information
	Image::Pixel pixel;

	for (int x = 0; x < width; x++)
		for (int y = 0; y < height; y++)
		{
			// setting pixel components
			pixel.i.r = x;
			pixel.i.g = 0;
			pixel.i.b = 0;
			pixel.i.a = 195;
			
			// setting the color of the image pixel at (x,y)
			image->set2D(x, y, pixel);
		}

	// setting our image as an albedo texture of the ground_grid material		
	MaterialPtr material = Materials::get()->findMaterial("ground_grid");
	material->setTextureImage(material->findTexture("albedo"), image);

	// saving our image to a *.DDS file
	image->save("myimage.dds");

	// printing current image format to the console
	Log::message("\nImage format: %s", image->getFormatName());

	// converting our image to the 32-bit float format
	image->convertToFormat(Image::FORMAT_RGBA32F);

	// printing image format after conversion to the console
	Log::message("\nImage format after conversion: %s", image->getFormatName());

	// getting the first pixel for each row of the image and printing its color value to the console
	for (int x = 0; x < width; x++)
	{
		pixel = image->get2D(x, 0);
		Log::message("\nPixel(%d, 0) color = (%f,%f,%f,%f)", x, pixel.f.r, pixel.f.g, pixel.f.b, pixel.f.a);
	}

	// clearing the image
	image->clear();
	
	return 1;
}

/* .. */

Image Class

Members

static ImagePtr create()

Default constructor. An empty instance with default parameters is created.

static ImagePtr create(const char * name)

Constructor. Creates an image loading data from the specified file.

Arguments

const char * name - Image file name.

static ImagePtr create(const Ptr<Image> & image)

Constructor. Creates an image by copying a given source image.

Arguments

const Ptr<Image> & image - Source image.

int set(int channel, int value)

Sets a value for a given image channel. Compressed and combined image formats are not supported.

Arguments

int channel - Channel number in range from 0 to number of channels - 1.
int value - Value in range from 0 to 255 (8-bit formats) or from 0 to 65535 (16-bit formats).

Return value

1 if the values is set successfully; otherwise, 0.

void set2D(int x, int y, const Image::Pixel & p)

Sets a color for a given pixel of a 2D image. Compressed formats are not supported.

Arguments

int x - X integer coordinate of a pixel in range [0; width].
int y - Y integer coordinate of a pixel in range [0; height].
const Image::Pixel & p - Pixel color represented by a Pixel structure.

Image::Pixel get2D(int x, int y)

Returns a color of a given pixel of a 2D image. Compressed formats are not supported.

Arguments

int x - X integer coordinate of a pixel in range [0; width].
int y - Y integer coordinate of a pixel in range [0; height].

Return value

Pixel color represented by a Pixel structure.

Image::Pixel get2D(float x, float y)

Returns a color of a given pixel of a 2D image. Linear interpolation is used to get a color sample. Compressed formats are not supported.

Arguments

float x - X float coordinate of a pixel in range [0.0f; 1.0f].
float y - Y float coordinate of a pixel in range [0.0f; 1.0f].

Return value

Pixel color represented by a Pixel structure.

Image::Pixel get2D(float x, float y, const Image::Pixel & skip_pixel)

Returns a color of a given pixel of a 2D image. Linear interpolation is used to get a color sample. Compressed formats are not supported.

NoticeIf at least one of four interpolated pixels is equal to skip_pixel, the method will return skip_pixel.

Arguments

float x - X float coordinate of a pixel in range [0.0f; 1.0f].
float y - Y float coordinate of a pixel in range [0.0f; 1.0f].
const Image::Pixel & skip_pixel - Skip pixel color represented by a Pixel structure.

Return value

Pixel color represented by a Pixel structure.

void set2DArray(int x, int y, int layer, const Image::Pixel & p)

Sets a color of a given pixel. The image must be a 2D image array. Compressed formats are not supported.

Arguments

int x - X integer coordinate of a pixel in range [0; width].
int y - Y integer coordinate of a pixel in range [0; height].
int layer - Image layer number.
const Image::Pixel & p - Pixel color represented by a Pixel structure.

Image::Pixel get2DArray(int x, int y, int layer)

Returns a color of a given pixel. The image must be a 2D texture array. Compressed formats are not supported.

Arguments

int x - X integer coordinate of a pixel in range [0; width].
int y - Y integer coordinate of a pixel in range [0; height].
int layer - Image layer number.

Return value

Pixel color represented by a Pixel structure.

Image::Pixel get2DArray(float x, float y, int layer)

Returns a color of a given pixel. The image must be a 2D image array. Compressed formats are not supported. Linear interpolation is used to get a color sample.

Arguments

float x - X float coordinate of a pixel in range [0.0f; 1.0f].
float y - Y float coordinate of a pixel in range [0.0f; 1.0f].
int layer - Image layer number.

Return value

Pixel color represented by a Pixel structure.

int is2DType()

Returns a value indicating if the image is a 2D image.

Return value

1 if the image is a 2D image; otherwise - 0.

void set3D(int x, int y, int z, const Image::Pixel & p)

Sets a color of a given pixel of a 3D image. Compressed formats are not supported.

Arguments

int x - X integer coordinate of a pixel in range [0; width].
int y - Y integer coordinate of a pixel in range [0; height].
int z - Z integer coordinate of a pixel in range [0; depth].
const Image::Pixel & p - Pixel color represented by a Pixel structure.

Image::Pixel get3D(float x, float y, float z)

Returns a color of a given pixel of a 3D image. Linear interpolation is used to get a color sample. Compressed formats are not supported.

Arguments

float x - X float coordinate of a pixel in range [0.0f; 1.0f].
float y - Y float coordinate of a pixel in range [0.0f; 1.0f].
float z - Z float coordinate of a pixel in range [0.0f; 1.0f].

Return value

Pixel color represented by a Pixel structure.

Image::Pixel get3D(int x, int y, int z)

Returns a color of a given pixel of a 3D image. Compressed formats are not supported.

Arguments

int x - X integer coordinate of a pixel in range [0; width].
int y - Y integer coordinate of a pixel in range [0; height].
int z - Z integer coordinate of a pixel in range [0; depth].

Return value

Pixel color represented by a Pixel structure.

Image::Pixel get3DSmooth(float x, float y, float z)

Returns a color of a given pixel obtained using linear interpolation. Compressed formats are not supported.

NoticeThis method returns pixel color with float components in the range [0.0f, 1.0f] regardless of the image format.

Arguments

float x - X float coordinate of a pixel in range [0.0f; 1.0f].
float y - Y float coordinate of a pixel in range [0.0f; 1.0f].
float z - Z float coordinate of a pixel in range [0.0f; 1.0f].

Return value

Pixel color represented by a Pixel structure with float components in the range [0.0f, 1.0f].

int is3DType()

Returns a value indicating if the image is a 3D image.

Return value

1 if the image is a 3D image; otherwise, 0.

int isArrayType()

Returns a value indicating if the image is of an array type (see *_ARRAY variables).

Return value

1 if the image is of an array type; otherwise, 0.

int isBCCompressedFormat()

Returns a value indicating if the image is compressed via GPU-accelerated DXT compression (DXT1, DXT1A, DXT3, DXT5, ATI1, ATI2 formats).

Return value

1 if the image is BC compressed; otherwise, 0.

int isBCCompressedFormat(int format)

Returns a value indicating if a given format is a format using GPU-accelerated DXT compression (DXT1, DXT1A, DXT3, DXT5, ATI1 or ATI2).

Arguments

int format - Format identifier. See the FORMAT_* variables.

Return value

1 if the specified format is a format using GPU-accelerated DXT compression; otherwise, 0.

int getBlockSize()

Returns the compressed image block size.

Return value

Block size in bytes.

int isCombinedFormat()

Returns a value indicating if the image is stored in a combined format (RGB565, RGBA4, RGB5A1 or RGB10A2).

Return value

1 if the image is in combined format; otherwise, 0.

int isCombinedFormat(int format)

Returns a value indicating if a given format is a combined format (RGB565, RGBA4, RGB5A1 or RGB10A2).

Arguments

int format - Format identifier. See the FORMAT_* variables.

Return value

1 if the specified format is a combined format; otherwise, 0.

int isCompressedFormat()

Returns a value indicating if the image is stored in a compressed format (DXT1, DXT3, DXT5, ATI1, ATI2, ZLC1 or ZLC2).

Return value

1 if the image is in a compressed format; otherwise, 0.

int isCompressedFormat(int format)

Returns a value indicating if a given format is a compressed format (DXT1, DXT3, DXT5, ATI1, ATI2, ZLC1 or ZLC2).

Arguments

int format - Format identifier. See the FORMAT_* variables.

Return value

1 if the specified format is a compressed format; otherwise, 0.

void setCube(int x, int y, int face, const Image::Pixel & p)

Sets a color of a given pixel. The image must be of the Cube type. Compressed formats are not supported.

Arguments

int x - X integer coordinate of a pixel in range [0; depth].
int y - Y integer coordinate of a pixel in range [0; height].
int face - Face number in range from 0 to 5.
const Image::Pixel & p - Pixel color represented by a Pixel structure.

Image::Pixel getCube(const Math::vec3 & direction)

Returns a color of a given pixel. The image must be of the Cube type. Compressed formats are not supported.

Arguments

const Math::vec3 & direction - Cube direction vector used for color sampling.

Return value

Pixel color represented by a Pixel structure.

Image::Pixel getCube(float x, float y, int face)

Returns a color of a given pixel. The image must be of the Cube type. Compressed formats are not supported.

Arguments

float x - X float coordinate of a pixel in range [0.0f; 1.0f].
float y - Y float coordinate of a pixel in range [0.0f; 1.0f].
int face - Face number in range from 0 to 5.

Return value

Pixel color represented by a Pixel structure.

Image::Pixel getCube(int x, int y, int face)

Returns a color of a given pixel. The image must be of the Cube type. Compressed formats are not supported.

Arguments

int x - X integer coordinate of a pixel in range [0; width].
int y - Y integer coordinate of a pixel in range [0; height].
int face - Face number in range from 0 to 5.

Return value

Pixel color represented by a Pixel structure.

void setCubeArray(int x, int y, int face, int layer, const Image::Pixel & p)

Sets a color of a given pixel. The image must be of the Cube Array type. Compressed formats are not supported.

Arguments

int x - X integer coordinate of a pixel in range [0; width].
int y - Y integer coordinate of a pixel in range [0; height].
int face - Face number in range from 0 to 5.
int layer - Image layer number.
const Image::Pixel & p - Pixel color represented by a Pixel structure.

Image::Pixel getCubeArray(const Math::vec3 & direction, int layer)

Returns the color of a given pixel of the cube array image.

Notice

Compressed formats are not supported.

Arguments

const Math::vec3 & direction - Cube direction vector used for color sampling.
int layer - Image layer number.

Return value

Pixel color represented by a Pixel structure.

Image::Pixel getCubeArray(int v1, int v2, int face, int layer)

Returns a color of a given pixel. The image must be of the Cube Array type. Compressed formats are not supported.

Arguments

int v1 - X integer coordinate of a pixel in range [0;width].
int v2 - Y integer coordinate of a pixel in range [0; height].
int face - Face number in range from 0 to 5.
int layer - Image layer number.

Return value

Pixel color represented by a Pixel structure.

Image::Pixel getCubeArray(float v1, float v2, int face, int layer)

Returns a color of a given pixel. The image must be of the Cube Array type. Compressed formats are not supported.

Arguments

float v1 - X integer coordinate of a pixel in range [0;width].
float v2 - Y integer coordinate of a pixel in range [0;height].
int face - Face number in range from 0 to 5.
int layer - Image layer number.

Return value

Pixel color represented by a Pixel structure.

int isCubeType()

Returns a value indicating if the image is Cube image.

Return value

1 if the image is a cube image; otherwise, 0.

int getDepth(int level = 0)

Returns the image depth on a given mipmap level.

Arguments

int level - Mipmap level in range from 0 to the total number of mipmaps. This is an optional argument.

Return value

Image depth if the image exists (its depth is larger than 0); otherwise, 1.

int isFloatFormat()

Returns a value indicating if the image is stored in a float format (R32F, RG32F, RGB32F or RGBA32F).

Return value

1 if the image is in float format; otherwise, 0 is returned.

int isFloatFormat(int format)

Returns a value indicating if a given format is a float format (R32F, RG32F, RGB32F or RGBA32F).

Arguments

int format - Format identifier. See the FORMAT_* variables.

Return value

1 if the specified format is a float format; otherwise, 0.

int getFormat()

Returns a numerical code of the image format.

Return value

Numerical code of the image format. See the FORMAT_* variables.

const char * getFormatName()

Returns an image format name. For more details about image formats see the article on ImageDDS tool.

Return value

Image format name. The following values can be returned: R8, RG8, RGB8, RGBA8, R16, RG16, RGB16, RGBA16, R16F, RG16F, RGB16F, RGBA16F, R32F, RG32F, RGB32F, RGBA32F, RGB565, RGBA4, RGB5A1, RGB10A2, DXT1, DXT3, DXT5, ATI1, ATI2, ZLC1, ZLC2.

int isHalfFormat()

Returns a value indicating if the image is stored in a half-float format (R16F, RG16F, RGB16F or RGBA16F).

Return value

1 if the image is in half-float format; otherwise, 0.

int isHalfFormat(int format)

Returns a value indicating if a given format is a half-float format (R16F, RG16F, RGB16F or RGBA16F).

Arguments

int format - Format identifier. See the FORMAT_* variables.

Return value

1 if the specified format is a half-float format; otherwise, 0.

int getHeight(int level = 0)

Returns image height on a given mipmap level.

Arguments

int level - Mipmap level in range from 0 to the total number of mipmaps. This is an optional argument.

Return value

Image height if the image exists (its height is larger than 0); otherwise, 1.

int isLoaded()

Returns a value indicating whether the image is loaded.

Return value

1 if the image is loaded; otherwise, 0.

int getNumChannels()

Returns the number of image channels.

Return value

Number of image channels.

int getNumFaces()

Returns the number of faces in the image.

Return value

Number of image faces.

int getNumLayers()

Returns the number of layers in the image.

Return value

Number of layers.

int getNumMipmaps()

Returns the number of mipmaps stored with the image.

Return value

Number of mipmap levels.

size_t getOffset(int level)

Returns the image offset on a given mipmap level.

Arguments

int level - Mipmap level in range from 0 to the total number of mipmaps.

Return value

Image offset, in bytes.

int isOwner()

Returns the owner flag. If the pointer is the owner, on its deletion the image also will be deleted. Use grab() and release() functions to change ownership.

Return value

Owner flag.

void setPixels(unsigned char * pixels)

Sets image pixels by copying data from a given source array.

Arguments

unsigned char * pixels - Source array.

void setPixels(const Ptr<Blob> & blob)

Adds the image to the blob.

Arguments

const Ptr<Blob> & blob - The image to be added.

unsigned char * getPixels()

Returns the pointer to the array of pixels.

Return value

Pointer to the array of pixels.

void getPixels(const Ptr<Blob> & blob)

Returns the pointer to the array of pixels.

Arguments

const Ptr<Blob> & blob - The target image.

unsigned char * getPixels2D(int level = 0)

Returns the pointer to the array of pixels with the given mipmap level for the 2D image.

Arguments

int level - Mipmap level in range from 0 to the total number of mipmaps. This is an optional argument.

Return value

Pointer to the array of pixels.

unsigned char * getPixels2DArray(int layer, int level = 0)

Returns the pointer to the array of pixels with the given mipmap level and image layer for the 2D Array image.

Arguments

int layer - Number of the image layer.
int level - Mipmap level in range from 0 to the total number of mipmaps. This is an optional argument.

Return value

Pointer to the array of pixels.

unsigned char * getPixels3D(int level = 0)

Returns the pointer to the array of pixels with the given mipmap level for the 3D image.

Arguments

int level - Mipmap level in range from 0 to the total number of mipmaps. This is an optional argument.

Return value

Pointer to the array of pixels.

unsigned char * getPixelsCube(int face, int level = 0)

Returns the pointer to the array of pixels with the given mipmap level for the cube image.

Arguments

int face - Face number in range from 0 to 5.
int level - Mipmap level in range from 0 to the total number of mipmaps. This is an optional argument.

Return value

Pointer to the array of pixels.

unsigned char * getPixelsCubeArray(int face, int layer, int level = 0)

Returns the pointer to the array of pixels with the given mipmap level and image layer for the Cube Array image.

Arguments

int face - Face number in range from 0 to 5.
int layer - Number of the image layer.
int level - Mipmap level in range from 0 to the total number of mipmaps. This is an optional argument.

Return value

Pointer to the array of pixels.

int getPixelSize()

Returns the uncompressed image pixel size.

Return value

Pixel size, in bytes.

int isRawFormat()

Returns a value indicating if the image is uncompressed (stored in R8, RG8, RGB8, RGBA8, R16, RG16, RGB16, RGBA16, R16F, RG16F, RGB16F, RGBA16F, R32F, RG32F, RGB32F, RGBA32F, RGB565, RGBA4, RGB5A1 or RGB10A2 format).

Return value

1 if the image is uncompressed; otherwise, 0.

int isRawFormat(int format)

Returns a value indicating if a given format is an uncompressed format (R8, RG8, RGB8, RGBA8, R16, RG16, RGB16, RGBA16, R16F, RG16F, RGB16F, RGBA16F, R32F, RG32F, RGB32F, RGBA32F, RGB565, RGBA4, RGB5A1 or RGB10A2).

Arguments

int format - Format identifier. See the FORMAT_* variables.

Return value

1 if the specified format is an uncompressed format; otherwise, 0.

size_t getSize()

Returns the current image size.

Return value

Image size, in bytes.

size_t getSize(int level)

Returns the current image size.

Arguments

int level - Mipmap level in range from 0 to the total number of mipmaps.

Return value

Image size in bytes.

size_t getStride(int level = 0)

Returns the image stride on a given mipmap level.

Arguments

int level - Mipmap level in range from 0 to the total number of mipmaps.

Return value

Image stride in bytes.

int getType()

Returns a numerical code of the image type.

Return value

Numerical code of the image type: 0 for a 2D image, 1 for a 3D image, 2 for a cube image, 3 for a 2D Array. See the Unigine::Image:: Enumeration with IMAGE_* prefixes.

const char * getTypeName()

Returns an image type name.

Return value

Image type name: 2D for a 2D image, 3D for a 3D image, Cube for a cube image, 2D_Array for a 2D Array.

int isUCharFormat()

Returns a value indicating if the image is stored in an unsigned char format (R8, RG8, RGB8, RGBA8, DXT1, DXT3, DXT5, ATI1 or ATI2).

Return value

1 if the image is in unsigned char format; otherwise, 0.

int isUCharFormat(int format)

Returns a value indicating if a given format is an unsigned char format (R8, RG8, RGB8, RGBA8, DXT1, DXT3, DXT5, ATI1 or ATI2).

Arguments

int format - Format identifier. See the FORMAT_* variables.

Return value

1 if the specified format is an unsigned char format; otherwise, 0.

int isUShortFormat()

Returns a value indicating if the image is stored in an unsigned short format (R16, RG16, RGB16 or RGBA16).

Return value

1 if the image is in unsigned short format; otherwise, 0.

int isUShortFormat(int format)

Returns a value indicating if a given format is an unsigned short format (R16, RG16, RGB16 or RGBA16).

Arguments

int format - Format identifier. See the FORMAT_* variables.

Return value

1 if the specified format is an unsigned short format; otherwise, 0.

int getWidth(int level = 0)

Returns the image width on a given mipmap level.

Arguments

int level - Mipmap level in range from 0 to the total number of mipmaps. This is an optional argument.

Return value

Image width if the image exists (its width is larger than 0); otherwise, 1.

int isZCompressedFormat()

Returns a value indicating if the image is stored in a ZLC1 or ZLC2 format (by ZLIB Image Compression).

Return value

1 if the image is in a z compressed format; otherwise, 0.

int isZCompressedFormat(int format)

Returns a value indicating if a given format is a format using ZLIB image compression (ZLC1 or ZLC2).

Arguments

int format - Format identifier. See the FORMAT_* variables.

Return value

1 if the specified format is a format using ZLIB image compression; otherwise, 0.

int blend(const Ptr<Image> & image, int x0, int y0, int x1, int y1, int width, int height, float scale = 1.0f, int safe = 0)

Blends the specified image with the current one. Blending takes place within a specified region. If the safe flag is set to 1, rendering of the blended images won't be performed outside the destination image boundaries. Compressed, combined, half-float and float formats are not supported. Images of different formats can be blended as follows:

R8 with R8, RG8;
RG8 with RG8, RGB8;
RGB8 with RGB8, RGBA8;
RGBA8 with RGBA8 only;
R16 with R16, RG16;
RG16 with RG16, RGB16;
RGB16 with RGB16, RGBA16;
RGBA16 with RGBA16 only;

Arguments

const Ptr<Image> & image - Source image to be blended with the current image.
int x0 - X coordinate of a position in the current image.
int y0 - Y coordinate of a position in the current image.
int x1 - X coordinate of a position in a source image, which is blended with the current image.
int y1 - Y coordinate of a position in a source image, which is blended with the current image.
int width - Width of the blending region.
int height - Height of the blending region.
float scale - Scale of blending:
- If blended images are of the same format, the blended image is multiplied by scale value and is added to the destination image.
- If a source image has one more channel compared to the current one, it will serve as alpha value. This channel value is multiplied by scale and is used to alpha blend the images.
- The default scale value is 1.0f.
int safe - 1 for safe blending with respect to destination image boundaries; otherwise, 0.

Return value

Returns 1 if blending is successful; otherwise, 0.

int blur(int size)

Filters the image using two-pass Gaussian blur. Only the 2D or cube image can be resized. Compressed and combined image formats are not supported.

Arguments

int size - Blur filter size. It is the number of neighbouring source pixels (from all sides) that are averaged for each blurred destination pixel. For example, if blur = 2, two pixels from each side are taken and the blurring kernel is 5 x 5 pixels.

Return value

1 if the image was successfully blurred or the size of blur filter is 0; otherwise, 0.

void clear()

Clears all data associated with the image and resets its type, format, size to default. Also the number of image layers and mipmaps is set to 1.

int combine(int new_format = -1)

Converts the image to a combined format. The following conversions are available:

RGB8 to RGB565
RGBA8 to RGBA4 if the format is specified as RGBA4; otherwise, RGBA8 is converted to RGB5A1 by default
RGBA16 to RGB10A2

Arguments

int new_format - Combined format to convert the image into. This is an optional argument. If no format is specified, default conversion will be performed.

Return value

1 if the conversion into the combined format is successful; otherwise, 0.

int combineMipmaps()

Combines the loaded 2D image with the mipmap image. Only one mipmap image can be combined. Compressed image formats cannot be combined.

Return value

1 if the image and a mipmap are successfully combined; otherwise, 0.

int compare(const Ptr<Image> & image, int x0, int y0, int x1, int y1, int width, int height, int threshold = 0)

Compares a region of a specified image with a specified region of the current image. Compressed, combined, half-float and float formats are not supported.

Arguments

const Ptr<Image> & image - Source image to compare the loaded image with.
int x0 - X coordinate on the loaded image.
int y0 - Y coordinate on the loaded image.
int x1 - X coordinate on the source image.
int y1 - Y coordinate on the source image.
int width - Width of a region to compare.
int height - Height of a region to compare.
int threshold - A threshold to compare.

Return value

1 if the regions match; otherwise, 0.

int compress(int new_format = -1)

Converts the image to a compressed format. The following conversions are available:

R8 to ATI1
RG8 to ATI2
RGB8 to DXT1
RGBA8 to DXT3, DXT5(by default)
ZLC1 to ZLC1
ZLC2 to ZLC2

Arguments

int new_format - Format to compress the image into. This is an optional argument. If no format is specified, default conversion will be performed.

Return value

1 if the image is compressed successfully; otherwise, 0.

int convertToFormat(int new_format)

Converts the image to a specified format. The following variants of conversion are possible:

Source type	Target type
DXT1	DXT3, RGBA8, DXT5, ZLC1, ZLC2
DXT3	DXT1, RGBA8 ZLC1, ZLC2
DXT5	DXT1, RGBA8, ZLC1, ZLC2
ATI1	R8, ATI2, ZLC1, ZLC2
ATI2	RG8, ATI1, ZLC1, ZLC2
RGB565	RGB8, DXT1, ZLC1, ZLC2
RGBA4	RGBA8, DXT3, DXT5, ZLC1, ZLC2
RGB5A1	RGBA8, DXT3, DXT5, ZLC1, ZLC2
RGB10A2	RGBA16, ZLC1, ZLC2
R8	ATI1, RG8, RGB8, RGBA8, R16, RG16, RGBA16, R32F, ATI2, DXT1, DXT3, DXT5, ZLC1, ZLC2
RG8	ATI2, R8, RGB8, RGBA8, R16, RG16, RGBA16, RG32F, DXT1, DXT3, DXT5, ZLC1, ZLC2
RGB8	R5G6B5, DXT1, R8, RG8, RGBA8, R16, RG16, RGBA16, RGB32F, ATI1, ATI2, DXT3, DXT5, ZLC1, ZLC2
RGBA8	RGBA4, RGB5A1, R8, RG8, RGB8, R16, RG16, RGB16, RGBA16, RGBA32F, ATI1, ATI2, DXT1, DXT3, DXT5, ZLC1, ZLC2
R16	R8, RG8, RGB8, RGBA8, RG16, RGBA16, R32F, ATI1, ATI2, DXT1, DXT3, DXT5, ZLC1, ZLC2
RG16	R8, RG8, RGB8, RGBA8, R16, RGBA16, RG32F, ATI1, ATI2, DXT1, DXT3, DXT5, ZLC1, ZLC2
RGB16	R8, RG8, RGB8, RGBA8, R16, RG16, RGBA16, RGB32F, ATI1, ATI2, DXT1, DXT3, DXT5, ZLC1, ZLC2
RGBA16	RGB10A2, R8, RG8, RGB8, RGBA8, R16, RG16, RGBA32F, ATI1, ATI2, DXT1, DXT3, DXT5, ZLC1, ZLC2
R16F	RG16F, RGB16F, RGBA16F, R32F, RG32F, RGB32F, RGBA32F, ZLC1, ZLC2
RG16F	R16F, RGB16F, RGBA16F, R32F, RG32F, RGB32F, RGBA32F, ZLC1, ZLC2
RGB16F	R16F, RG16F, RGBA16F, R32F, RG32F, RGB32F, RGBA32F, ZLC1, ZLC2
RGBA16F	R16F, RG16F, RGB16F, R32F, RG32F, RGB32F, RGBA32F, ZLC1, ZLC2
R32F	R8, R16, R16F, RG16F, RGB16F, RGBA16F, RG32F, RGB32F, RGBA32F, ATI1, ZLC1, ZLC2
RG32F	RG8, RG16, R16F, RG16F, RGB16F, RGBA16F, R32F, RGB32F, RGBA32F, ATI2, ZLC1, ZLC2
RGB32F	RGB8, RGB16, R16F, RG16F, RGB16F, RGBA16F, R32F, RG32F, RGB32F, RGBA32F, DXT1, ZLC1, ZLC2
RGBA32F	RGBA8, RGBA16, R16F, RG16F, RGB16F, RGBA16F, R32F, RG32F, RGB32F, DXT3, DXT5, ZLC1, ZLC2

Arguments

int new_format - Target format to convert the image into. See the FORMAT_* variables.

Return value

1 if the conversion is successful; otherwise, 0.

int convertToType(int type)

Converts the image to a specified type. The following conversions are possible:

Source type	Target type	Target dimensions
2D (height should be proportional to width)	3D	width × height × height/width
2D (height should divide by four, width should divide by three)	Cube	width/4 × height/3
3D	2D	width × height*depth
3D	2D texture array	width × height
Cube	2D	width4 × height3
Cube	2D array texture	width × height
2D array texture	2D	width × height*number of texture layers

Arguments

int type - Type for conversion. One of the IMAGE_* types.

Return value

1 if the conversion is successful; otherwise, 0.

int copy(const Ptr<Image> & src_image, int layer)

Copies a specified layer between images.

Arguments

const Ptr<Image> & src_image - Source image.
int layer - Layer number.

Return value

1 if the layer is copied successfully; otherwise, 0.

int copy(const Ptr<Image> & src_image, int x_dst, int y_dst, int x_src, int y_src, int width, int height, int safe = 0)

Copies a given layer between images.

Arguments

const Ptr<Image> & src_image - Source image, from which a region will be copied.
int x_dst - X coordinate of a position in the image, where data will be copied to.
int y_dst - Y coordinate of a position in the image, where data will be copied to.
int x_src - X coordinate of a position in a source image, where data will be copied from.
int y_src - Y coordinate of a position in a source image, where data will be copied from.
int width - Width of an image region to copy.
int height - Height of an image region to copy.
int safe - 1 for safe copying with respect to destination image boundaries; otherwise, 0.
Notice
The safe flag doesn't provide safe copying with respect to source image boundaries.

Return value

1 if the layer is copied successfully; otherwise, 0.

int copy(const Ptr<Image> & src_image, int dest_channel, int src_channel)

Copies a given layer between images.

Arguments

const Ptr<Image> & src_image - Source image.
int dest_channel - Destination channel.
int src_channel - Source channel.

Return value

1 if the layer is copied successfully; otherwise, 0.

int create2D(int width, int height, int format, int num_mipmaps = 1, int clear = 1, int allocate = 1)

Creates an empty 2D image filled with black color.

Arguments

int width - Required width.
int height - Required height.
int format - Required format. See the FORMAT_* variables.
int num_mipmaps - Number of mipmaps to create. This is an optional argument. If the value is smaller than 1, an assertion is raised.
int clear - Positive number to fill the image with black color, 0 not to fill. This optional argument can be set to 0 if new data will fully overwrite the data existing in the memory.
int allocate - Positive number to allocate memory for the created image. This value should be always set to 1.

Return value

1 if 2D image is successfully created; otherwise, 0.

int create2DArray(int width, int height, int num_layers, int format, int num_mipmaps = 1, int clear = 1, int allocate = 1)

Creates an empty 2D texture array filled with black color.

Arguments

int width - Required width.
int height - Required height.
int num_layers - Number of texture layers.
int format - Required format. See the FORMAT_* variables.
int num_mipmaps - Number of mipmaps to create. This is an optional argument. If the value is smaller than 1, an assertion is raised.
int clear - Positive number to fill the texture array with black color, 0 not to fill. This optional argument can be set to 0 if new data will fully overwrite the data existing in the memory.
int allocate - Positive number to allocate memory for the created image. This value should be always set to 1.

Return value

1 if 2D texture array is successfully created; otherwise, 0.

int create3D(int width, int height, int depth, int format, int num_mipmaps = 1, int clear = 1, int allocate = 1)

Creates an empty 3D image filled with black color.

Arguments

int width - Required width.
int height - Required height.
int depth - Required depth.
int format - Required format. See the IMAGE_FORMAT_* variables.
int num_mipmaps - Number of mipmaps to create. This is an optional argument. If the value is smaller than 1, an assertion is raised.
int clear - Positive number to fill the image with black color, 0 not to fill. This optional argument can be set to 0 if new data will fully overwrite the data existing in the memory.
int allocate - Positive number to allocate memory for the created image. This value should be always set to 1.

Return value

1 if 3D image is successfully created; otherwise, 0.

int createCube(int width, int height, int format, int num_mipmaps = 1, int clear = 1, int allocate = 1)

Creates the empty cube image filled with black color.

Arguments

int width - Image width, in pixels.
int height - Image height, in pixels.
int format - Texture format. See the FORMAT_* variables.
int num_mipmaps - Number of mipmaps to create. This is an optional argument.
int clear - 1 to fill the image with black color, 0 is not to fill. This optional argument can be set to 0 if new data will fully overwrite the data existing in the memory.
int allocate - 1 to allocate memory for the created image. This value should be always set to 1.

Return value

1 if the cube image is successfully created; otherwise, 0.

int createCubeArray(int width, int height, int num_layers, int format, int num_mipmaps = 1, int clear = 1, int allocate = 1)

Creates the empty cube array image filled with black color.

Arguments

int width - Image width, in pixels.
int height - Image height, in pixels.
int num_layers - Number of image layers.
int format - Texture format. See the FORMAT_* variables.
int num_mipmaps - Number of mipmaps to create. This is an optional argument.
int clear - 1 to fill the image with black color, 0 is not to fill. This optional argument can be set to 0 if new data will fully overwrite the data existing in the memory.
int allocate - 1 to allocate memory for the created image. This value should be always set to 1.

Return value

1 if the cube array image is successfully created; otherwise, 0.

int createMipmaps(int filter = FILTER_LINEAR, float gamma = 1.0f)

Generates mipmaps for the image. Mipmaps cannot be created for compressed and combined image formats.

Arguments

int filter - Filter type to be used: box or point. The default filter is box. See the Unigine::Image Enumerations with FILTER_* prefixes at the end of the article.
float gamma - Gamma correction value. If 1 is provided, no gamma correction is applied. This is an optional argument.

Return value

1 if the mipmaps are generated successfully; otherwise, 0.

int decombine()

Automatically converts the image from the combined format to a correct one. The following conversions are available:

RGB565 to RGB8
RGBA4 to RGBA8
RGB5A1 to RGBA8
RGB10A2 to RGBA16

Return value

1 if the image is successfully converted from the combined format; otherwise, 0.

int decombineMipmaps()

Decombines the loaded 2D-image and the mipmap images. The number of mipmaps must be larger than 1. Compressed image formats cannot be decombined.

Return value

1 if the 2D-image and mipmap image are successfully decombined; otherwise, 0.

int decompress()

Decompresses the image from the compressed format to a correct one. The following conversions are available:

DXT1 to RGB8
DXT1 without alpha data to RGB8
DXT3 to RGBA8
DXT5 to RGBA8
ATI1 to R8
ATI2 to RG8

Return value

1 if the image was successfully decompressed; otherwise, 0.

int extrude(int pixels)

Extrudes contour of the image in the alpha=0 region on the given number of pixels. Only images of the RGBA8 and RGBA16 formats can be extruded.

Arguments

int pixels - Number of contour pixels.

Return value

1 if the image contour is successfully extruded; otherwise, 0.

int flipX()

Flips the 2D image horizontally. Compressed and combined image formats are not supported.

Return value

int flipY()

Flips the 2D image vertically. Compressed and combined image formats are not supported.

Return value

1 if the image is successfully flipped; otherwise, 0.

void grab()

Grabs the image (sets the owner flag to 1). The image should not be handled by the engine after this function is called.

int hasMipmaps()

Returns a value indicating if the source image has mipmaps.

Return value

1 if the image has mipmaps; otherwise, 0.

int info(const char * name)

Returns information about the given image.

Arguments

const char * name - Name of the image file. The following file formats are supported: TGA, JPG, PNG, DDS.

Return value

1 if the operation was a success; otherwise, 0.

int invert()

Inverts values of all the image channels. Only R8, RG8, RGB8, RGBA8, R16, RG16, RGB16, RGBA16, R32F, RG32F, RGB32F and RGBA32F formats can be inverted.

Return value

1 if the image is inverted successfully; otherwise, 0.

int load(const char * file, int offset)

Loads an image from a specified file.

Arguments

const char * file - Name of the image file. The following file formats are supported: JPG and PNG.
int offset - Mipmap offset value for the loaded image.

Return value

1 if the image is loaded successfully; otherwise, 0.

int load(const char * file)

Loads an image from a specified file.

Arguments

const char * file - Name of the image file. The following file formats are supported: JPG and PNG.

Return value

1 if the image is loaded successfully; otherwise, 0.

int load(const Ptr<Stream> & stream)

Loads an image from a specified stream.

Arguments

const Ptr<Stream> & stream - Stream spart pointer.

Return value

1 if the image is loaded successfully; otherwise, 0.

int normalize()

Creates a three-component normal map out of the image.

Return value

1 if normalization was successful; otherwise, 0.

void release()

Releases the image (sets the owner flag to 0). The image should be handled by the engine after this function is called.

int resize(int new_width, int new_height, int filter = FILTER_LINEAR)

Resizes the image. Only the 2D or cube image can be resized. Compressed and combined image formats cannot be resized.

Arguments

int new_width - New image width.
int new_height - New image height.
int filter - Quality setting, accepts one of the FILTER_* values.

Return value

1 if the image is resized successfully; otherwise, 0.

int rotate(int angle)

Rotates the image by a given angle (with the step of 90 degrees). Only the 2D or cube image can be rotated. Compressed and combined image formats cannot be rotated.

Arguments

int angle - Angle that is multiplied by the step (angle * 90) to set the image rotation degree.

Return value

1 if the image is rotated successfully; otherwise, 0.

int save(const char * name)

Saves the image to a specified file. The following file formats are supported:

Arguments

const char * name - Name of the file (with the specified extension) to save image data.

Return value

Returns 1 if the image is saved successfully; otherwise, 0.

int save(const char * name, float quality)

Saves the image to a specified file. The following file formats are supported:

Arguments

const char * name - Name of a file with extension into which data will be saved.
float quality - Quality of JPG compression in range [0;1].

Return value

Returns 1 if the image is saved successfully; otherwise, 0.

int save(const Ptr<Stream> & stream)

Saves the image to a stream. The following file formats are supported:

Arguments

const Ptr<Stream> & stream - Pointer to the stream.

Return value

Returns 1 if the image is saved successfully; otherwise, 0.

int sign()

Converts the image from unsigned type to signed char. Only R8, RG8, RGB8, RGBA8, R16, RG16, RGB16, RGBA16, ATI1 and ATI2 formats can be converted.

Return value

1 if the image has been converted successfully; otherwise, 0.

int swap(int channel, int index)

Swaps values of two channels. Compressed formats are not supported.

Arguments

int channel - First channel number in range of 0 to (number of channels - 1).
int index - Second channel number in range of 0 to (number of channels - 1).

Return value

1 if the values are swapped successfully; otherwise, 0.

void swap(const Ptr<Image> & image)

Swaps the data of the current mesh file with the source mesh file.

Arguments

const Ptr<Image> & image - Source mesh.

Image::Pixel toPixel(const Math::vec4 & color)

Converts a vec4 color vector (R, G, B, A) to Pixel color.

Arguments

const Math::vec4 & color - Color vec4 vector (R, G, B, A) with components normalized in range [0.0f; 1.0f]

Return value

Pixel color represented by a Pixel structure.

Math::vec4 toVec4(const Image::Pixel & pixel)

Converts Pixel color to a vec4 color vector (R, G, B, A).

Arguments

const Image::Pixel & pixel - Pixel color represented by a Pixel structure.

Return value

Color vec4 vector (R, G, B, A). Each component is normalized in range [0.0f; 1.0f]

int FILTER_LINEAR

Description

A linear mipmap filter.

int FILTER_POINT

Description

A point mipmap filter.

int FORMAT_ATI1

Description

An ATI1 compressed image file format. Compresses single-channel images by a factor of 2:1 compared to R8. Compression is the same as the one used for alpha data in the DXT5 format. Supported by all graphics cards compatible with DirectX 10; additionally, all ATI graphics cards since X800.

int FORMAT_ATI2

Description

An ATI2 compressed image file format. Compresses two-channel images by a factor of 2:1 compared to RG8. Compression for each channel is the same as the one used for alpha data in the DXT5 format. Supported by all graphics cards compatible with DirectX 10; additionally, all ATI graphics cards since X800.

int FORMAT_DXT1

Description

A DXT1 compressed image file format. Compresses RGB images by a factor of 6:1 compared to RGB8. Uses 4×4 blocks for color interpolation and stores color values in 16 bits per pixel (R5:G6:B5). Can also store 1-bit alpha data; in this case, compresses images by a factor of 8:1 compared to RGBA8. This format is not recommended for storing normal maps.

int FORMAT_DXT3

Description

A DXT3 compressed image file format. Compresses RGBA images by a factor of 4:1 compared to RGBA8. Stores RGB color values as the DXT1 format and uses four more bits for alpha data. Color values are interpreted as not having been pre-multiplied by alpha. Alpha values are not interpolated. Recommended for images with sharp alpha transitions.

int FORMAT_DXT5

Description

A DXT5 compressed image file format. Compresses RGBA images by a factor of 4:1 compared to RGBA8. Stores RGB color values as the DXT1 format and uses four more bits for alpha data. Color values are interpreted as not having been pre-multiplied by alpha. Alpha values are interpolated. Recommended for images with smooth alpha transitions.

int FORMAT_LZ4

Description

An LZ4 lossless data compression algorithm that is focused on compression and decompression speed. It belongs to the LZ77 family of byte-oriented compression schemes.

int FORMAT_R16

Description

Non-compressed unsigned format. Stores data in one 16-bit channel (red).

int FORMAT_R16F

Description

Non-compressed float format. Stores data in one 16-bit channel (red).

int FORMAT_R32F

Description

Non-compressed float format. Stores data in one 32-bit channel (red).

int FORMAT_R8

Description

Non-compressed unsigned format. Stores data in one 8-bit channel (red).

int FORMAT_RG16

Description

Non-compressed unsigned format. Stores data in two 16-bit channels (red, green).

int FORMAT_RG16F

Description

Non-compressed float format. Stores data in two 16-bit channels (red, green).

int FORMAT_RG32F

Description

Non-compressed float format. Stores data in two 32-bit channels (red, green).

int FORMAT_RG8

Description

Non-compressed unsigned format. Stores data in two 8-bit channels (red, green).

int FORMAT_RGB10A2

Description

Combined unsigned format. Stores data in three 10-bit channels (red, green, blue) and one 2-bit channel (alpha).

int FORMAT_RGB16

Description

Non-compressed unsigned format. Stores data in three 16-bit channels (red, green, blue).

int FORMAT_RGB16F

Description

Non-compressed float format. Stores data in three 16-bit channels (red, green, blue).

int FORMAT_RGB32F

Description

Non-compressed float format. Stores data in three 32-bit channels (red, green, blue).

int FORMAT_RGB565

Description

Combined unsigned format. Stores 16 bit color values without alpha.

int FORMAT_RGB5A1

Description

Combined unsigned format. Stores data in three 5-bit channels (red, green, blue) and one 1-bit channel (alpha).

int FORMAT_RGB8

Description

Non-compressed unsigned format. Stores data in three 8-bit channels (red, green, blue).

int FORMAT_RGBA16

Description

Non-compressed unsigned format. Stores data in four 16-bit channels (red, green, blue, alpha).

int FORMAT_RGBA16F

Description

Non-compressed float format. Stores data in four 16-bit channels (red, green, blue, alpha).

int FORMAT_RGBA32F

Description

Non-compressed float format. Stores data in four 32-bit channels (red, green, blue, alpha).

int FORMAT_RGBA4

Description

Combined unsigned format. Stores data in four 4-bit channels (red, green, blue, alpha).

int FORMAT_RGBA8

Description

Non-compressed unsigned format. Stores data in four 8-bit channels (red, green, blue, alpha).

int FORMAT_ZLC1

Description

Compressed format used for images stored in the system memory (based on zlib library). It is used for fast compression (for example, in run-time).

int FORMAT_ZLC2

Description

Compressed format used for images stored in the system memory (based on zlib library). It is used when a smaller file size is required, if compared to ZLC1 format.

int IMAGE_2D

Description

A 2D texture.

int IMAGE_2D_ARRAY

Description

A 2D texture array. Each texture of the array has its own mipmap levels, which is the difference between it and a 3D image.

int IMAGE_3D

Description

A 3D texture.

int IMAGE_CUBE

Description

A cube texture.

int IMAGE_CUBE_ARRAY

Description

A cube texture array.

Notice

Not supported on OpenGLES 3.0.

Last update: 2017-10-20

Help improve this article

Was this article helpful?

Report a problem related to this article

(or select a word/phrase and press Ctrl+Enter)

Unigine::Image Class

List of functions:

List of Variables:

Usage example

Image Class

Members

static ImagePtr create()

static ImagePtr create(const char * name)

Arguments

static ImagePtr create(const Ptr<Image> & image)

Arguments

int set(int channel, int value)

Arguments

Return value

void set2D(int x, int y, const Image::Pixel & p)

Arguments

Image::Pixel get2D(int x, int y)

Arguments

Return value

Image::Pixel get2D(float x, float y)

Arguments

Return value

Image::Pixel get2D(float x, float y, const Image::Pixel & skip_pixel)

Arguments

Return value

void set2DArray(int x, int y, int layer, const Image::Pixel & p)

Arguments

Image::Pixel get2DArray(int x, int y, int layer)

Arguments

Return value

Image::Pixel get2DArray(float x, float y, int layer)

Arguments

Return value

int is2DType()

Return value

void set3D(int x, int y, int z, const Image::Pixel & p)

Arguments

Image::Pixel get3D(float x, float y, float z)

Arguments

Return value

Image::Pixel get3D(int x, int y, int z)

Arguments

Return value

Image::Pixel get3DSmooth(float x, float y, float z)

Arguments

Return value

int is3DType()

Return value

int isArrayType()

Return value

int isBCCompressedFormat()

Return value

int isBCCompressedFormat(int format)

Arguments

Return value

int getBlockSize()

Return value

int isCombinedFormat()

Return value

int isCombinedFormat(int format)

Arguments

Return value

int isCompressedFormat()

Return value

int isCompressedFormat(int format)

Arguments

Return value

void setCube(int x, int y, int face, const Image::Pixel & p)

Arguments

Image::Pixel getCube(const Math::vec3 & direction)

Arguments

Return value

Image::Pixel getCube(float x, float y, int face)

Arguments

Return value

Image::Pixel getCube(int x, int y, int face)

Arguments

Return value

void setCubeArray(int x, int y, int face, int layer, const Image::Pixel & p)

Arguments