boofcv.alg.distort

Class DistortImageOps

java.lang.Object

boofcv.alg.distort.DistortImageOps

public class DistortImageOps
extends Object

Provides common function for distorting images.

Author:

Peter Abeles

Constructor Summary

Constructors

Constructor and Description
DistortImageOps()

Method Summary

Methods

Modifier and Type	Method and Description
static <T extends ImageBase> void	affine(T input, T output, TypeInterpolate interpType, double a11, double a12, double a21, double a22, double dx, double dy) Applies an affine transformation from the input image to the output image.
static georegression.struct.shapes.Rectangle2D_F32	boundBox_F32(int srcWidth, int srcHeight, PixelTransform_F32 transform) Finds an axis-aligned bounding box which would contain a image after it has been transformed.
static georegression.struct.shapes.Rectangle2D_I32	boundBox(int srcWidth, int srcHeight, int dstWidth, int dstHeight, PixelTransform_F32 transform) Finds an axis-aligned bounding box which would contain a image after it has been transformed.
static georegression.struct.shapes.Rectangle2D_I32	boundBox(int srcWidth, int srcHeight, PixelTransform_F32 transform) Finds an axis-aligned bounding box which would contain a image after it has been transformed.
static <T extends ImageSingleBand> ImageDistort<T>	createImageDistort(PointTransform_F32 transform, TypeInterpolate interpType, Class<T> imageType) Easy way to create ImageDistort given PixelTransform_F32.
static <T extends ImageSingleBand,M extends MultiSpectral<T>> void	distortMS(M input, M output, PixelTransform_F32 transform, boolean skipOutsidePixels, TypeInterpolate interpType) Applies a pixel transform to a MultiSpectral image.
static <T extends ImageSingleBand> void	distortMS(MultiSpectral<T> input, MultiSpectral<T> output, ImageDistort<T> distortion) Applies a distortion to a MultiSpectral image.
static <T extends ImageSingleBand> void	distortSingle(T input, T output, PixelTransform_F32 transform, boolean skipOutsidePixels, TypeInterpolate interpType) Applies a pixel transform to a single band image.
static <T extends ImageSingleBand> void	distortSingle(T input, T output, PixelTransform_F32 transform, ImageBorder<T> border, InterpolatePixelS<T> interp) Applies a pixel transform to a single band image.
static <T extends ImageBase> void	rotate(T input, T output, TypeInterpolate interpType, float angleInputToOutput) Rotates the image using the specified interpolation type.
static <T extends ImageBase> void	scale(T input, T output, TypeInterpolate interpType) Rescales the input image and writes the results into the output image.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

DistortImageOps

public DistortImageOps()

Method Detail

affine

public static <T extends ImageBase> void affine(T input,
                                T output,
                                TypeInterpolate interpType,
                                double a11,
                                double a12,
                                double a21,
                                double a22,
                                double dx,
                                double dy)

Applies an affine transformation from the input image to the output image.

Input coordinates (x,y) to output coordinate (x',y')
x' = a11*x + a12*y + dx
y' = a21*x + a22*y + dy

Parameters:

input - Which which is being rotated.

output - The image in which the output is written to.

interpType - Which type of interpolation will be used.

distortSingle

public static <T extends ImageSingleBand> void distortSingle(T input,
                                             T output,
                                             PixelTransform_F32 transform,
                                             boolean skipOutsidePixels,
                                             TypeInterpolate interpType)

Applies a pixel transform to a single band image. Easier to use function.

Parameters:

input - Input (source) image.

output - Where the result of transforming the image image is written to.

transform - The transform that is being applied to the image

skipOutsidePixels - Should pixels that go outside the source image be skipped over or set to the value of zero.

interpType - Which type of pixel interpolation should be used. BILINEAR is in general recommended

distortSingle

public static <T extends ImageSingleBand> void distortSingle(T input,
                                             T output,
                                             PixelTransform_F32 transform,
                                             ImageBorder<T> border,
                                             InterpolatePixelS<T> interp)

Applies a pixel transform to a single band image. More flexible but order to use function.

Parameters:

input - Input (source) image.

output - Where the result of transforming the image image is written to.

transform - The transform that is being applied to the image

border - How border pixels are handled.

interp - Interpolation algorithm.

distortMS

public static <T extends ImageSingleBand,M extends MultiSpectral<T>> void distortMS(M input,
                                                                    M output,
                                                                    PixelTransform_F32 transform,
                                                                    boolean skipOutsidePixels,
                                                                    TypeInterpolate interpType)

Applies a pixel transform to a MultiSpectral image.

Parameters:

input - Input (source) image.

output - Where the result of transforming the image image is written to.

transform - The transform that is being applied to the image

skipOutsidePixels - Should pixels that go outside the source image be skipped over or set to the value of zero.

interpType - Which type of pixel interpolation should be used.

createImageDistort

public static <T extends ImageSingleBand> ImageDistort<T> createImageDistort(PointTransform_F32 transform,
                                                             TypeInterpolate interpType,
                                                             Class<T> imageType)

Easy way to create ImageDistort given PixelTransform_F32. To improve performance the distortion is automatically cached.

Parameters:

transform - Image transform.

interpType - Which interpolation. Try bilinear.

imageType - Image of single band image it will process.

Returns:

The ImageDistort

See Also:

FactoryDistort, FactoryInterpolation

scale

public static <T extends ImageBase> void scale(T input,
                               T output,
                               TypeInterpolate interpType)

Rescales the input image and writes the results into the output image. The scale factor is determined independently of the width and height.

Parameters:

input - Input image. Not modified.

output - Rescaled input image. Modified.

interpType - Which interpolation algorithm should be used.

rotate

public static <T extends ImageBase> void rotate(T input,
                                T output,
                                TypeInterpolate interpType,
                                float angleInputToOutput)

Rotates the image using the specified interpolation type. The rotation is performed around the specified center of rotation in the input image.

Input coordinates (x,y) to output coordinate (x',y')
x' = x_c + c*(x-x_c) - s(y - y_c)
y' = y_c + s*(x-x_c) + c(y - y_c)

Parameters:

input - Which which is being rotated.

output - The image in which the output is written to.

interpType - Which type of interpolation will be used.

angleInputToOutput - Angle of rotation in radians. From input to output, CCW rotation.

distortMS

public static <T extends ImageSingleBand> void distortMS(MultiSpectral<T> input,
                                         MultiSpectral<T> output,
                                         ImageDistort<T> distortion)

Applies a distortion to a MultiSpectral image.

Type Parameters:

T - Band type.

Parameters:

input - Image being distorted. Not modified.

output - Output image. modified.

distortion - The distortion model

boundBox

public static georegression.struct.shapes.Rectangle2D_I32 boundBox(int srcWidth,
                                                   int srcHeight,
                                                   int dstWidth,
                                                   int dstHeight,
                                                   PixelTransform_F32 transform)

Finds an axis-aligned bounding box which would contain a image after it has been transformed. A sanity check is done to made sure it is contained inside the destination image's bounds. If it is totally outside then a rectangle with negative width or height is returned.

Parameters:

srcWidth - Width of the source image

srcHeight - Height of the source image

dstWidth - Width of the destination image

dstHeight - Height of the destination image

transform - Transform being applied to the image

Returns:

Bounding box

boundBox

public static georegression.struct.shapes.Rectangle2D_I32 boundBox(int srcWidth,
                                                   int srcHeight,
                                                   PixelTransform_F32 transform)

Finds an axis-aligned bounding box which would contain a image after it has been transformed. The returned bounding box can be larger then the original image.

Parameters:

srcWidth - Width of the source image

srcHeight - Height of the source image

transform - Transform being applied to the image

Returns:

Bounding box

boundBox_F32

public static georegression.struct.shapes.Rectangle2D_F32 boundBox_F32(int srcWidth,
                                                       int srcHeight,
                                                       PixelTransform_F32 transform)

Finds an axis-aligned bounding box which would contain a image after it has been transformed. The returned bounding box can be larger then the original image.

Parameters:

srcWidth - Width of the source image

srcHeight - Height of the source image

transform - Transform being applied to the image

Returns:

Bounding box