com.emphysic.myriad.core.data.ops

Class CannyOperation

java.lang.Object

com.emphysic.myriad.core.data.ops.CannyOperation

All Implemented Interfaces:

DatasetOperation, ObjectMap, com.esotericsoftware.kryo.KryoSerializable

public class CannyOperation
extends java.lang.Object
implements DatasetOperation

CannyOperation - Canny Edge detection Created by chris on 8/25/2016.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	CannyOperation.AutoThreshold Whether and how to attempt automatic calculation of the lower and upper thresholds.

Field Summary

Fields

Modifier and Type	Field and Description
(package private) int[]	cardinals Cardinal points - horizontal, vertical, and diagonal.

Constructor Summary

Constructors

Constructor and Description
CannyOperation() Constructs a new Canny Edge detector, denoising input with a Gaussian Blur of radius 5 and calculating gradients with a standard Sobel edge detector.
CannyOperation(BlurOperation blurOperation, SobelOperation convolutionOperation) Constructs a new Canny Edge detector with the specified denoise operation and gradient operation
CannyOperation(BlurOperation blurOperation, SobelOperation convolutionOperation, CannyOperation.AutoThreshold autoThreshold) Constructs a new Canny Edge detector with the specified denoise operation and gradient operation
CannyOperation(int blurRadius) Constructs a new Canny Edge detector, denoising input with a Gaussian Blur of the specified radius and calculating gradients with a standard Sobel edge detector.

Method Summary

Modifier and Type	Method and Description
CannyOperation.AutoThreshold	getAutoThreshold() Returns the current settings for auto thresholding
BlurOperation	getBlurOperation() Returns the smoothing operation used to denoise input data.
SobelOperation	getGradientOperation() Returns the operation used to calculate horizontal and vertical gradients.
double	getLowerThreshold() Gets the "weak" or lower threshold for edge detection - elements with values less than this value are not considered edges and are rejected.
java.util.Map<java.lang.String,java.lang.Object>	getObjectMap() Creates a map of the important fields for the instance, suitable for serialization.
long	getSerializationVersion() Returns the current version of the serialization format.
double	getSigmaThreshold() Returns the current threshold parameter for auto-thresholding based on the mean or median.
double	getUpperThreshold() Gets the "strong" or upper threshold for edge detection - elements with values greater than or equal to this value are considered edges.
int	getVersion() Returns the current class version.
double	gradMax(Dataset input, int i, int j, double angle) Calculates the local maximum along the gradient.
Dataset	hysteresis(Dataset input) Calculates the "hysteresis thresholding" of the input.
void	initCurrentVersion(java.util.Map<java.lang.String,java.lang.Object> objectMap) Initializes an instance with a current-version object graph.
double	roundToCardinal(double radians) Rounds the specified angle in radians to the closest cardinal direction
Dataset	run(Dataset input) Runs the operation on an input.
void	setAutoThreshold(CannyOperation.AutoThreshold autoThreshold) Sets the auto thresholding
void	setBlurOperation(BlurOperation blurOperation) Sets the operation used to denoise data prior to edge detection.
void	setGradientOperation(SobelOperation gradientOperation) Sets the operation used to calculate horizontal and vertical gradients
void	setLowerThreshold(double lowerThreshold) Sets the "weak" or lower threshold for edge detection - elements with values greater than or equal to this value are considered edges.
void	setSigmaThreshold(double sigmaThreshold) Sets the threshold parameter for auto-thresholding based on the mean or median
void	setUpperThreshold(double upperThreshold) Sets the "strong" or upper threshold for edge detection - elements with values greater than or equal to this value are considered edges.

Methods inherited from class java.lang.Object

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

Methods inherited from interface com.emphysic.myriad.core.data.util.ObjectMap

init, initPreviousVersion, initUnknownVersion, load, read, save, write

Field Detail

cardinals

int[] cardinals

Cardinal points - horizontal, vertical, and diagonal. Local gradient directions are rounded to one of these directions: 0, 45, 90, 135 (i.e. horizontal, vertical, or diagonal).

Constructor Detail

CannyOperation

public CannyOperation(BlurOperation blurOperation,
                      SobelOperation convolutionOperation,
                      CannyOperation.AutoThreshold autoThreshold)

Constructs a new Canny Edge detector with the specified denoise operation and gradient operation

Parameters:

blurOperation - BlurOperation used to smooth and denoise input data

convolutionOperation - ConvolutionOperation used to calculate horizontal and vertical gradients

autoThreshold - whether/how to automatically determine thresholds

CannyOperation

public CannyOperation(BlurOperation blurOperation,
                      SobelOperation convolutionOperation)

Constructs a new Canny Edge detector with the specified denoise operation and gradient operation

Parameters:

blurOperation - BlurOperation used to smooth and denoise input data

convolutionOperation - ConvolutionOperation used to calculate horizontal and vertical gradients

CannyOperation

public CannyOperation()

Constructs a new Canny Edge detector, denoising input with a Gaussian Blur of radius 5 and calculating gradients with a standard Sobel edge detector.

CannyOperation

public CannyOperation(int blurRadius)

Constructs a new Canny Edge detector, denoising input with a Gaussian Blur of the specified radius and calculating gradients with a standard Sobel edge detector.

Parameters:

blurRadius - radius of smoothing operation

Method Detail

run

public Dataset run(Dataset input)

Description copied from interface: DatasetOperation

Runs the operation on an input.

Specified by:

run in interface DatasetOperation

Parameters:

input - Dataset on which to operate

Returns:

result

hysteresis

public Dataset hysteresis(Dataset input)

Calculates the "hysteresis thresholding" of the input. For each element in the input: 1. If element > upper threshold, it's an edge: set to 1 2. If element < lower threshold, it's a reject: set to 0 3. Otherwise it's a candidate: if at least one of its 8 neighbors is an edge it's also an edge, otherwise reject

Parameters:

input - Dataset to threshold

Returns:

thresholded Dataset : edges set to 1, all else to 0

roundToCardinal

public double roundToCardinal(double radians)

Rounds the specified angle in radians to the closest cardinal direction

Parameters:

radians - angle in radians

Returns:

closest angle in degrees: one of 0, 45, 90, or 135

gradMax

public double gradMax(Dataset input,
                      int i,
                      int j,
                      double angle)

Calculates the local maximum along the gradient. For a point (i, j) under consideration, find the perpendicular to the gradient angle at (i, j). Take the two points closes to (i, j) that lie on this perpendicular, and calculate the maximum of the three points. Used to thin edges - keep the largest value and scrap the others.

Parameters:

input - input dataset

i - center element's horizontal index within input

j - center element's vertical index within input

angle - angle of gradient at point i, j

Returns:

local maximum value of the three points along a line perpendicular to the angle.

getBlurOperation

public BlurOperation getBlurOperation()

Returns the smoothing operation used to denoise input data.

Returns:

smoothing operation

setBlurOperation

public void setBlurOperation(BlurOperation blurOperation)

Sets the operation used to denoise data prior to edge detection.

Parameters:

blurOperation - smoothing operation

getGradientOperation

public SobelOperation getGradientOperation()

Returns the operation used to calculate horizontal and vertical gradients.

Returns:

gradient operation

setGradientOperation

public void setGradientOperation(SobelOperation gradientOperation)

Sets the operation used to calculate horizontal and vertical gradients

Parameters:

gradientOperation - gradient operation to use

getUpperThreshold

public double getUpperThreshold()

Gets the "strong" or upper threshold for edge detection - elements with values greater than or equal to this value are considered edges.

Returns:

upper threshold

setUpperThreshold

public void setUpperThreshold(double upperThreshold)

Sets the "strong" or upper threshold for edge detection - elements with values greater than or equal to this value are considered edges. Typically 2-3 times the size of the lower threshold.

Parameters:

upperThreshold - strong threshold value

getLowerThreshold

public double getLowerThreshold()

Gets the "weak" or lower threshold for edge detection - elements with values less than this value are not considered edges and are rejected.

Returns:

lower threshold

setLowerThreshold

public void setLowerThreshold(double lowerThreshold)

Sets the "weak" or lower threshold for edge detection - elements with values greater than or equal to this value are considered edges. Typically the upper threshold is 2-3 times the size of the lower threshold.

Parameters:

lowerThreshold - weak threshold value

getAutoThreshold

public CannyOperation.AutoThreshold getAutoThreshold()

Returns the current settings for auto thresholding

Returns:

AutoThreshold enum

setAutoThreshold

public void setAutoThreshold(CannyOperation.AutoThreshold autoThreshold)

Sets the auto thresholding

Parameters:

autoThreshold - new thresholding setting

getSigmaThreshold

public double getSigmaThreshold()

Returns the current threshold parameter for auto-thresholding based on the mean or median.

Returns:

threshold parameter (0-1)

setSigmaThreshold

public void setSigmaThreshold(double sigmaThreshold)

Sets the threshold parameter for auto-thresholding based on the mean or median

Parameters:

sigmaThreshold - parameter between 0 and 1

getSerializationVersion

public long getSerializationVersion()

Description copied from interface: ObjectMap

Returns the current version of the serialization format. Increments when the serialization format changes, i.e. backwards compatibility is broken.

Specified by:

getSerializationVersion in interface ObjectMap

Returns:

serialization version

getVersion

public int getVersion()

Description copied from interface: ObjectMap

Returns the current class version. Class version increments when the class changes but doesn't necessarily break compatibility with serialization.

Specified by:

getVersion in interface ObjectMap

Returns:

class version

getObjectMap

public java.util.Map<java.lang.String,java.lang.Object> getObjectMap()

Description copied from interface: ObjectMap

Creates a map of the important fields for the instance, suitable for serialization.

Specified by:

getObjectMap in interface ObjectMap

Returns:

"object map" of the instance

initCurrentVersion

public void initCurrentVersion(java.util.Map<java.lang.String,java.lang.Object> objectMap)

Description copied from interface: ObjectMap

Initializes an instance with a current-version object graph.

Specified by:

initCurrentVersion in interface ObjectMap

Parameters:

objectMap - object graph for initialization