org.nd4j.autodiff.samediff.ops

Class SDImage

java.lang.Object

org.nd4j.autodiff.samediff.ops.SDOps

org.nd4j.autodiff.samediff.ops.SDImage

public class SDImage
extends SDOps

Field Summary

Fields inherited from class org.nd4j.autodiff.samediff.ops.SDOps

sd

Constructor Summary

Constructors

Constructor and Description
SDImage(SameDiff sameDiff)

Method Summary

Modifier and Type	Method and Description
SDVariable	adjustContrast(SDVariable in, double factor) Adjusts contrast of RGB or grayscale images.
SDVariable	adjustContrast(String name, SDVariable in, double factor) Adjusts contrast of RGB or grayscale images.
SDVariable	adjustHue(SDVariable in, double delta) Adjust hue of RGB image
SDVariable	adjustHue(String name, SDVariable in, double delta) Adjust hue of RGB image
SDVariable	adjustSaturation(SDVariable in, double factor) Adjust saturation of RGB images
SDVariable	adjustSaturation(String name, SDVariable in, double factor) Adjust saturation of RGB images
SDVariable	cropAndResize(SDVariable image, SDVariable cropBoxes, SDVariable boxIndices, SDVariable cropOutSize) Given an input image and some crop boxes, extract out the image subsets and resize them to the specified size.
SDVariable	cropAndResize(SDVariable image, SDVariable cropBoxes, SDVariable boxIndices, SDVariable cropOutSize, double extrapolationValue) Given an input image and some crop boxes, extract out the image subsets and resize them to the specified size.
SDVariable	cropAndResize(String name, SDVariable image, SDVariable cropBoxes, SDVariable boxIndices, SDVariable cropOutSize) Given an input image and some crop boxes, extract out the image subsets and resize them to the specified size.
SDVariable	cropAndResize(String name, SDVariable image, SDVariable cropBoxes, SDVariable boxIndices, SDVariable cropOutSize, double extrapolationValue) Given an input image and some crop boxes, extract out the image subsets and resize them to the specified size.
SDVariable	extractImagePatches(SDVariable image, int[] kSizes, int[] strides, int[] rates, boolean sameMode) Given an input image, extract out image patches (of size kSizes - h x w) and place them in the depth dimension.
SDVariable	extractImagePatches(String name, SDVariable image, int[] kSizes, int[] strides, int[] rates, boolean sameMode) Given an input image, extract out image patches (of size kSizes - h x w) and place them in the depth dimension.
SDVariable	hsvToRgb(SDVariable input) Converting image from HSV to RGB format
SDVariable	hsvToRgb(String name, SDVariable input) Converting image from HSV to RGB format
SDVariable	imageResize(SDVariable input, SDVariable size, boolean preserveAspectRatio, boolean antialis, ImageResizeMethod ImageResizeMethod) Resize images to size using the specified method.
SDVariable	imageResize(SDVariable input, SDVariable size, ImageResizeMethod ImageResizeMethod) Resize images to size using the specified method.
SDVariable	imageResize(String name, SDVariable input, SDVariable size, boolean preserveAspectRatio, boolean antialis, ImageResizeMethod ImageResizeMethod) Resize images to size using the specified method.
SDVariable	imageResize(String name, SDVariable input, SDVariable size, ImageResizeMethod ImageResizeMethod) Resize images to size using the specified method.
SDVariable	nonMaxSuppression(SDVariable boxes, SDVariable scores, int maxOutSize, double iouThreshold, double scoreThreshold) Greedily selects a subset of bounding boxes in descending order of score
SDVariable	nonMaxSuppression(String name, SDVariable boxes, SDVariable scores, int maxOutSize, double iouThreshold, double scoreThreshold) Greedily selects a subset of bounding boxes in descending order of score
SDVariable	randomCrop(SDVariable input, SDVariable shape) Randomly crops image
SDVariable	randomCrop(String name, SDVariable input, SDVariable shape) Randomly crops image
SDVariable	rgbToHsv(SDVariable input) Converting array from HSV to RGB format
SDVariable	rgbToHsv(String name, SDVariable input) Converting array from HSV to RGB format
SDVariable	rgbToYiq(SDVariable input) Converting array from RGB to YIQ format
SDVariable	rgbToYiq(String name, SDVariable input) Converting array from RGB to YIQ format
SDVariable	rgbToYuv(SDVariable input) Converting array from RGB to YUV format
SDVariable	rgbToYuv(String name, SDVariable input) Converting array from RGB to YUV format
SDVariable	yiqToRgb(SDVariable input) Converting image from YIQ to RGB format
SDVariable	yiqToRgb(String name, SDVariable input) Converting image from YIQ to RGB format
SDVariable	yuvToRgb(SDVariable input) Converting image from YUV to RGB format
SDVariable	yuvToRgb(String name, SDVariable input) Converting image from YUV to RGB format

Methods inherited from class java.lang.Object

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

Constructor Detail

SDImage

public SDImage(SameDiff sameDiff)

Method Detail

cropAndResize

public SDVariable cropAndResize(SDVariable image,
                                SDVariable cropBoxes,
                                SDVariable boxIndices,
                                SDVariable cropOutSize,
                                double extrapolationValue)

Given an input image and some crop boxes, extract out the image subsets and resize them to the specified size.

Parameters:

image - Input image, with shape [batch, height, width, channels] (NUMERIC type)

cropBoxes - Float32 crop, shape [numBoxes, 4] with values in range 0 to 1 (NUMERIC type)

boxIndices - Indices: which image (index to dimension 0) the cropBoxes belong to. Rank 1, shape [numBoxes] (NUMERIC type)

cropOutSize - Output size for the images - int32, rank 1 with values [outHeight, outWidth] (INT type)

extrapolationValue - Used for extrapolation, when applicable. 0.0 should be used for the default

Returns:

output Cropped and resized images (NUMERIC type)

cropAndResize

public SDVariable cropAndResize(String name,
                                SDVariable image,
                                SDVariable cropBoxes,
                                SDVariable boxIndices,
                                SDVariable cropOutSize,
                                double extrapolationValue)

Given an input image and some crop boxes, extract out the image subsets and resize them to the specified size.

Parameters:

name - name May be null. Name for the output variable

image - Input image, with shape [batch, height, width, channels] (NUMERIC type)

cropBoxes - Float32 crop, shape [numBoxes, 4] with values in range 0 to 1 (NUMERIC type)

boxIndices - Indices: which image (index to dimension 0) the cropBoxes belong to. Rank 1, shape [numBoxes] (NUMERIC type)

cropOutSize - Output size for the images - int32, rank 1 with values [outHeight, outWidth] (INT type)

extrapolationValue - Used for extrapolation, when applicable. 0.0 should be used for the default

Returns:

output Cropped and resized images (NUMERIC type)

cropAndResize

public SDVariable cropAndResize(SDVariable image,
                                SDVariable cropBoxes,
                                SDVariable boxIndices,
                                SDVariable cropOutSize)

Given an input image and some crop boxes, extract out the image subsets and resize them to the specified size.

Parameters:

image - Input image, with shape [batch, height, width, channels] (NUMERIC type)

cropBoxes - Float32 crop, shape [numBoxes, 4] with values in range 0 to 1 (NUMERIC type)

boxIndices - Indices: which image (index to dimension 0) the cropBoxes belong to. Rank 1, shape [numBoxes] (NUMERIC type)

cropOutSize - Output size for the images - int32, rank 1 with values [outHeight, outWidth] (INT type)

Returns:

output Cropped and resized images (NUMERIC type)

cropAndResize

public SDVariable cropAndResize(String name,
                                SDVariable image,
                                SDVariable cropBoxes,
                                SDVariable boxIndices,
                                SDVariable cropOutSize)

Given an input image and some crop boxes, extract out the image subsets and resize them to the specified size.

Parameters:

name - name May be null. Name for the output variable

image - Input image, with shape [batch, height, width, channels] (NUMERIC type)

cropBoxes - Float32 crop, shape [numBoxes, 4] with values in range 0 to 1 (NUMERIC type)

boxIndices - Indices: which image (index to dimension 0) the cropBoxes belong to. Rank 1, shape [numBoxes] (NUMERIC type)

cropOutSize - Output size for the images - int32, rank 1 with values [outHeight, outWidth] (INT type)

Returns:

output Cropped and resized images (NUMERIC type)

adjustContrast

public SDVariable adjustContrast(SDVariable in,
                                 double factor)

Adjusts contrast of RGB or grayscale images.

Parameters:

in - images to adjust. 3D shape or higher (NUMERIC type)

factor - multiplier for adjusting contrast

Returns:

output Contrast-adjusted image (NUMERIC type)

adjustContrast

public SDVariable adjustContrast(String name,
                                 SDVariable in,
                                 double factor)

Adjusts contrast of RGB or grayscale images.

Parameters:

name - name May be null. Name for the output variable

in - images to adjust. 3D shape or higher (NUMERIC type)

factor - multiplier for adjusting contrast

Returns:

output Contrast-adjusted image (NUMERIC type)

adjustHue

public SDVariable adjustHue(SDVariable in,
                            double delta)

Adjust hue of RGB image

Parameters:

in - image as 3D array (NUMERIC type)

delta - value to add to hue channel

Returns:

output adjusted image (NUMERIC type)

adjustHue

public SDVariable adjustHue(String name,
                            SDVariable in,
                            double delta)

Adjust hue of RGB image

Parameters:

name - name May be null. Name for the output variable

in - image as 3D array (NUMERIC type)

delta - value to add to hue channel

Returns:

output adjusted image (NUMERIC type)

adjustSaturation

public SDVariable adjustSaturation(SDVariable in,
                                   double factor)

Adjust saturation of RGB images

Parameters:

in - RGB image as 3D array (NUMERIC type)

factor - factor for saturation

Returns:

output adjusted image (NUMERIC type)

adjustSaturation

public SDVariable adjustSaturation(String name,
                                   SDVariable in,
                                   double factor)

Adjust saturation of RGB images

Parameters:

name - name May be null. Name for the output variable

in - RGB image as 3D array (NUMERIC type)

factor - factor for saturation

Returns:

output adjusted image (NUMERIC type)

extractImagePatches

public SDVariable extractImagePatches(SDVariable image,
                                      int[] kSizes,
                                      int[] strides,
                                      int[] rates,
                                      boolean sameMode)

Given an input image, extract out image patches (of size kSizes - h x w) and place them in the depth dimension.

Parameters:

image - Input image to extract image patches from - shape [batch, height, width, channels] (NUMERIC type)

kSizes - Kernel size - size of the image patches, [height, width] (Size: Exactly(count=2))

strides - Stride in the input dimension for extracting image patches, [stride_height, stride_width] (Size: Exactly(count=2))

rates - Usually [1,1]. Equivalent to dilation rate in dilated convolutions - how far apart the output pixels in the patches should be, in the input. A dilation of [a,b] means every ath pixel is taken along the height/rows dimension, and every bth pixel is take along the width/columns dimension (Size: AtLeast(min=0))

sameMode - Padding algorithm. If true: use Same padding

Returns:

output The extracted image patches (NUMERIC type)

extractImagePatches

public SDVariable extractImagePatches(String name,
                                      SDVariable image,
                                      int[] kSizes,
                                      int[] strides,
                                      int[] rates,
                                      boolean sameMode)

Given an input image, extract out image patches (of size kSizes - h x w) and place them in the depth dimension.

Parameters:

name - name May be null. Name for the output variable

image - Input image to extract image patches from - shape [batch, height, width, channels] (NUMERIC type)

kSizes - Kernel size - size of the image patches, [height, width] (Size: Exactly(count=2))

strides - Stride in the input dimension for extracting image patches, [stride_height, stride_width] (Size: Exactly(count=2))

rates - Usually [1,1]. Equivalent to dilation rate in dilated convolutions - how far apart the output pixels in the patches should be, in the input. A dilation of [a,b] means every ath pixel is taken along the height/rows dimension, and every bth pixel is take along the width/columns dimension (Size: AtLeast(min=0))

sameMode - Padding algorithm. If true: use Same padding

Returns:

output The extracted image patches (NUMERIC type)

hsvToRgb

public SDVariable hsvToRgb(SDVariable input)

Converting image from HSV to RGB format

Parameters:

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

hsvToRgb

public SDVariable hsvToRgb(String name,
                           SDVariable input)

Converting image from HSV to RGB format

Parameters:

name - name May be null. Name for the output variable

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

imageResize

public SDVariable imageResize(SDVariable input,
                              SDVariable size,
                              boolean preserveAspectRatio,
                              boolean antialis,
                              ImageResizeMethod ImageResizeMethod)

Resize images to size using the specified method.

Parameters:

input - 4D image [NHWC] (NUMERIC type)

size - new height and width (INT type)

preserveAspectRatio - Whether to preserve the aspect ratio. If this is set, then images will be resized to a size that fits in size while preserving the aspect ratio of the original image. Scales up the image if size is bigger than the current size of the image. Defaults to False.

antialis - Whether to use an anti-aliasing filter when downsampling an image

ImageResizeMethod - ResizeBilinear: Bilinear interpolation. If 'antialias' is true, becomes a hat/tent filter function with radius 1 when downsampling. ResizeLanczos5: Lanczos kernel with radius 5. Very-high-quality filter but may have stronger ringing. ResizeBicubic: Cubic interpolant of Keys. Equivalent to Catmull-Rom kernel. Reasonably good quality and faster than Lanczos3Kernel, particularly when upsampling. ResizeGaussian: Gaussian kernel with radius 3, sigma = 1.5 / 3.0. ResizeNearest: Nearest neighbor interpolation. 'antialias' has no effect when used with nearest neighbor interpolation. ResizeArea: Anti-aliased resampling with area interpolation. 'antialias' has no effect when used with area interpolation; it always anti-aliases. ResizeMitchelcubic: Mitchell-Netravali Cubic non-interpolating filter. For synthetic images (especially those lacking proper prefiltering), less ringing than Keys cubic kernel but less sharp.

Returns:

output Output image (NUMERIC type)

imageResize

public SDVariable imageResize(String name,
                              SDVariable input,
                              SDVariable size,
                              boolean preserveAspectRatio,
                              boolean antialis,
                              ImageResizeMethod ImageResizeMethod)

Resize images to size using the specified method.

Parameters:

name - name May be null. Name for the output variable

input - 4D image [NHWC] (NUMERIC type)

size - new height and width (INT type)

preserveAspectRatio - Whether to preserve the aspect ratio. If this is set, then images will be resized to a size that fits in size while preserving the aspect ratio of the original image. Scales up the image if size is bigger than the current size of the image. Defaults to False.

antialis - Whether to use an anti-aliasing filter when downsampling an image

ImageResizeMethod - ResizeBilinear: Bilinear interpolation. If 'antialias' is true, becomes a hat/tent filter function with radius 1 when downsampling. ResizeLanczos5: Lanczos kernel with radius 5. Very-high-quality filter but may have stronger ringing. ResizeBicubic: Cubic interpolant of Keys. Equivalent to Catmull-Rom kernel. Reasonably good quality and faster than Lanczos3Kernel, particularly when upsampling. ResizeGaussian: Gaussian kernel with radius 3, sigma = 1.5 / 3.0. ResizeNearest: Nearest neighbor interpolation. 'antialias' has no effect when used with nearest neighbor interpolation. ResizeArea: Anti-aliased resampling with area interpolation. 'antialias' has no effect when used with area interpolation; it always anti-aliases. ResizeMitchelcubic: Mitchell-Netravali Cubic non-interpolating filter. For synthetic images (especially those lacking proper prefiltering), less ringing than Keys cubic kernel but less sharp.

Returns:

output Output image (NUMERIC type)

imageResize

public SDVariable imageResize(SDVariable input,
                              SDVariable size,
                              ImageResizeMethod ImageResizeMethod)

Resize images to size using the specified method.

Parameters:

input - 4D image [NHWC] (NUMERIC type)

size - new height and width (INT type)

ImageResizeMethod - ResizeBilinear: Bilinear interpolation. If 'antialias' is true, becomes a hat/tent filter function with radius 1 when downsampling. ResizeLanczos5: Lanczos kernel with radius 5. Very-high-quality filter but may have stronger ringing. ResizeBicubic: Cubic interpolant of Keys. Equivalent to Catmull-Rom kernel. Reasonably good quality and faster than Lanczos3Kernel, particularly when upsampling. ResizeGaussian: Gaussian kernel with radius 3, sigma = 1.5 / 3.0. ResizeNearest: Nearest neighbor interpolation. 'antialias' has no effect when used with nearest neighbor interpolation. ResizeArea: Anti-aliased resampling with area interpolation. 'antialias' has no effect when used with area interpolation; it always anti-aliases. ResizeMitchelcubic: Mitchell-Netravali Cubic non-interpolating filter. For synthetic images (especially those lacking proper prefiltering), less ringing than Keys cubic kernel but less sharp.

Returns:

output Output image (NUMERIC type)

imageResize

public SDVariable imageResize(String name,
                              SDVariable input,
                              SDVariable size,
                              ImageResizeMethod ImageResizeMethod)

Resize images to size using the specified method.

Parameters:

name - name May be null. Name for the output variable

input - 4D image [NHWC] (NUMERIC type)

size - new height and width (INT type)

ImageResizeMethod - ResizeBilinear: Bilinear interpolation. If 'antialias' is true, becomes a hat/tent filter function with radius 1 when downsampling. ResizeLanczos5: Lanczos kernel with radius 5. Very-high-quality filter but may have stronger ringing. ResizeBicubic: Cubic interpolant of Keys. Equivalent to Catmull-Rom kernel. Reasonably good quality and faster than Lanczos3Kernel, particularly when upsampling. ResizeGaussian: Gaussian kernel with radius 3, sigma = 1.5 / 3.0. ResizeNearest: Nearest neighbor interpolation. 'antialias' has no effect when used with nearest neighbor interpolation. ResizeArea: Anti-aliased resampling with area interpolation. 'antialias' has no effect when used with area interpolation; it always anti-aliases. ResizeMitchelcubic: Mitchell-Netravali Cubic non-interpolating filter. For synthetic images (especially those lacking proper prefiltering), less ringing than Keys cubic kernel but less sharp.

Returns:

output Output image (NUMERIC type)

nonMaxSuppression

public SDVariable nonMaxSuppression(SDVariable boxes,
                                    SDVariable scores,
                                    int maxOutSize,
                                    double iouThreshold,
                                    double scoreThreshold)

Greedily selects a subset of bounding boxes in descending order of score

Parameters:

boxes - Might be null. Name for the output variable (NUMERIC type)

scores - vector of shape [num_boxes] (NUMERIC type)

maxOutSize - scalar representing the maximum number of boxes to be selected

iouThreshold - threshold for deciding whether boxes overlap too much with respect to IOU

scoreThreshold - threshold for deciding when to remove boxes based on score

Returns:

output vectort of shape [M] representing the selected indices from the boxes tensor, where M <= max_output_size (NUMERIC type)

nonMaxSuppression

public SDVariable nonMaxSuppression(String name,
                                    SDVariable boxes,
                                    SDVariable scores,
                                    int maxOutSize,
                                    double iouThreshold,
                                    double scoreThreshold)

Greedily selects a subset of bounding boxes in descending order of score

Parameters:

name - name May be null. Name for the output variable

boxes - Might be null. Name for the output variable (NUMERIC type)

scores - vector of shape [num_boxes] (NUMERIC type)

maxOutSize - scalar representing the maximum number of boxes to be selected

iouThreshold - threshold for deciding whether boxes overlap too much with respect to IOU

scoreThreshold - threshold for deciding when to remove boxes based on score

Returns:

output vectort of shape [M] representing the selected indices from the boxes tensor, where M <= max_output_size (NUMERIC type)

randomCrop

public SDVariable randomCrop(SDVariable input,
                             SDVariable shape)

Randomly crops image

Parameters:

input - input array (NUMERIC type)

shape - shape for crop (INT type)

Returns:

output cropped array (NUMERIC type)

randomCrop

public SDVariable randomCrop(String name,
                             SDVariable input,
                             SDVariable shape)

Randomly crops image

Parameters:

name - name May be null. Name for the output variable

input - input array (NUMERIC type)

shape - shape for crop (INT type)

Returns:

output cropped array (NUMERIC type)

rgbToHsv

public SDVariable rgbToHsv(SDVariable input)

Converting array from HSV to RGB format

Parameters:

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

rgbToHsv

public SDVariable rgbToHsv(String name,
                           SDVariable input)

Converting array from HSV to RGB format

Parameters:

name - name May be null. Name for the output variable

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

rgbToYiq

public SDVariable rgbToYiq(SDVariable input)

Converting array from RGB to YIQ format

Parameters:

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

rgbToYiq

public SDVariable rgbToYiq(String name,
                           SDVariable input)

Converting array from RGB to YIQ format

Parameters:

name - name May be null. Name for the output variable

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

rgbToYuv

public SDVariable rgbToYuv(SDVariable input)

Converting array from RGB to YUV format

Parameters:

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

rgbToYuv

public SDVariable rgbToYuv(String name,
                           SDVariable input)

Converting array from RGB to YUV format

Parameters:

name - name May be null. Name for the output variable

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

yiqToRgb

public SDVariable yiqToRgb(SDVariable input)

Converting image from YIQ to RGB format

Parameters:

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

yiqToRgb

public SDVariable yiqToRgb(String name,
                           SDVariable input)

Converting image from YIQ to RGB format

Parameters:

name - name May be null. Name for the output variable

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

yuvToRgb

public SDVariable yuvToRgb(SDVariable input)

Converting image from YUV to RGB format

Parameters:

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)

yuvToRgb

public SDVariable yuvToRgb(String name,
                           SDVariable input)

Converting image from YUV to RGB format

Parameters:

name - name May be null. Name for the output variable

input - 3D image (NUMERIC type)

Returns:

output 3D image (NUMERIC type)