DenseSIFT (OpenIMAJ master project 1.3.10 API)

java.lang.Object
- org.openimaj.image.feature.dense.gradient.dsift.AbstractDenseSIFT<FImage>
- - org.openimaj.image.feature.dense.gradient.dsift.DenseSIFT

All Implemented Interfaces:

Cloneable, ImageAnalyser<FImage>

Direct Known Subclasses:

ApproximateDenseSIFT
```
public class DenseSIFT
extends AbstractDenseSIFT<FImage>
```
Implementation of a dense SIFT feature extractor for FImages. Extracts upright SIFT features at a single scale on a grid.
Implementation inspired by the VLFeat extractor.
Implementation Notes. The analyser is not thread-safe, however, it is safe to reuse the analyser. In multi-threaded environments, a separate instance must be made for each thread. Internally, this implementation allocates memory for the gradient images, and if possible re-uses these between calls. Re-use requires that the input image is the same size between calls to the analyser.

Author:

Jonathon Hare (jsh2@ecs.soton.ac.uk)

See Also:

"http://www.vlfeat.org/api/dsift.html#dsift-usage"

Field Summary

Fields
Modifier and Type	Field and Description
`protected int`	`binHeight` Height of a single bin of the sampling window (in pixels).
`protected int`	`binWidth` Width of a single bin of the sampling window (in pixels).
`protected org.openimaj.image.feature.dense.gradient.dsift.DenseSIFT.WorkingData`	`data`
`protected float[][]`	`descriptors` Extracted descriptors
`protected float[]`	`energies` Descriptor energies
`protected float`	`gaussianWindowSize` Size of the Gaussian window (in relative to of the size of a bin)
`protected int`	`numBinsX` Number of spatial bins in the X direction
`protected int`	`numBinsY` Number of spatial bins in the Y direction
`protected int`	`numOriBins` The number of orientation bins
`protected int`	`stepX` Step size of sampling window in x-direction (in pixels)
`protected int`	`stepY` Step size of sampling window in y-direction (in pixels)
`protected float`	`valueThreshold` Threshold for clipping the SIFT features

Constructor Summary

Constructors
Constructor and Description
`DenseSIFT()` Construct with the default configuration: standard SIFT geometry (4x4x8), 5px x 5px spatial bins, 5px step size, gaussian window size of 2 and value threshold of 0.2.
`DenseSIFT(int step, int binSize)` Construct with the given step size (for both x and y) and binSize.
`DenseSIFT(int stepX, int stepY, int binWidth, int binHeight, int numBinsX, int numBinsY, int numOriBins)` Construct with the given configuration.
`DenseSIFT(int stepX, int stepY, int binWidth, int binHeight, int numBinsX, int numBinsY, int numOriBins, float gaussianWindowSize)` Construct with the given configuration.
`DenseSIFT(int stepX, int stepY, int binWidth, int binHeight, int numBinsX, int numBinsY, int numOriBins, float gaussianWindowSize, float valueThreshold)` Construct with the given configuration.

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`void`	`analyseImage(FImage image, Rectangle bounds)` Compute the dense sift descriptors inside the bounds rectangle of the given image.
`DenseSIFT`	`clone()`
`protected void`	`extractFeatures()` Extract the DSIFT features
`int`	`getBinHeight()` (Optional operation) Get the height of a single bin of the sampling window (in pixels).
`int`	`getBinWidth()` (Optional operation) Get the width of a single bin of the sampling window (in pixels).
`LocalFeatureList<ByteDSIFTKeypoint>`	`getByteKeypoints()` Get the SIFT descriptors from the previous call to `AbstractDenseSIFT.analyseImage(Image)` or `AbstractDenseSIFT.analyseImage(Image, Rectangle)` in the form of a list of local features with byte vectors.
`LocalFeatureList<ByteDSIFTKeypoint>`	`getByteKeypoints(float energyThreshold)` Get the SIFT descriptors from the previous call to `AbstractDenseSIFT.analyseImage(Image)` or `AbstractDenseSIFT.analyseImage(Image, Rectangle)` in the form of a list of local features with byte vectors.
`float[][]`	`getDescriptors()` Get the computed raw dense SIFT descriptors from the previous call to `#analyseImage(FImage)` or `analyseImage(FImage, Rectangle)` .
`LocalFeatureList<FloatDSIFTKeypoint>`	`getFloatKeypoints()` Get the SIFT descriptors from the previous call to `AbstractDenseSIFT.analyseImage(Image)` or `AbstractDenseSIFT.analyseImage(Image, Rectangle)` in the form of a list of local features with float vectors.
`LocalFeatureList<FloatDSIFTKeypoint>`	`getFloatKeypoints(float energyThreshold)` Get the SIFT descriptors from the previous call to `AbstractDenseSIFT.analyseImage(Image)` or `AbstractDenseSIFT.analyseImage(Image, Rectangle)` in the form of a list of local features with float vectors.
`int`	`getNumBinsX()` Get the number of spatial bins in the X direction
`int`	`getNumBinsY()` Get the number of spatial bins in the Y direction
`int`	`getNumOriBins()` Get the number of orientation bins
`void`	`setBinHeight(int size)` (Optional operation) Set the height of a single bin of the sampling window (in pixels).
`void`	`setBinWidth(int size)` (Optional operation) Set the width of a single bin of the sampling window (in pixels).

Methods inherited from class org.openimaj.image.feature.dense.gradient.dsift.AbstractDenseSIFT
analyseImage

Methods inherited from class java.lang.Object
equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - stepX
```
protected int stepX
```
    Step size of sampling window in x-direction (in pixels)
  - stepY
```
protected int stepY
```
    Step size of sampling window in y-direction (in pixels)
  - binWidth
```
protected int binWidth
```
    Width of a single bin of the sampling window (in pixels). Sampling window width is this multiplied by #numBinX.
  - binHeight
```
protected int binHeight
```
    Height of a single bin of the sampling window (in pixels). Sampling window height is this multiplied by #numBinY.
  - numBinsX
```
protected int numBinsX
```
    Number of spatial bins in the X direction
  - numBinsY
```
protected int numBinsY
```
    Number of spatial bins in the Y direction
  - numOriBins
```
protected int numOriBins
```
    The number of orientation bins
  - gaussianWindowSize
```
protected float gaussianWindowSize
```
    Size of the Gaussian window (in relative to of the size of a bin)
  - valueThreshold
```
protected float valueThreshold
```
    Threshold for clipping the SIFT features
  - data
```
protected volatile org.openimaj.image.feature.dense.gradient.dsift.DenseSIFT.WorkingData data
```
  - descriptors
```
protected volatile float[][] descriptors
```
    Extracted descriptors
  - energies
```
protected volatile float[] energies
```
    Descriptor energies
- Constructor Detail
  - DenseSIFT
```
public DenseSIFT()
```
    Construct with the default configuration: standard SIFT geometry (4x4x8), 5px x 5px spatial bins, 5px step size, gaussian window size of 2 and value threshold of 0.2.
  - DenseSIFT
```
public DenseSIFT(int step,
                 int binSize)
```
    Construct with the given step size (for both x and y) and binSize. All other values are the defaults.
    
    Parameters:
    
    step - the step size
    
    binSize - the spatial bin size
  - DenseSIFT
```
public DenseSIFT(int stepX,
                 int stepY,
                 int binWidth,
                 int binHeight,
                 int numBinsX,
                 int numBinsY,
                 int numOriBins)
```
    Construct with the given configuration. The gaussian window size is set to 2, and value threshold to 0.2.
    
    Parameters:
    
    stepX - step size in x direction
    
    stepY - step size in y direction
    
    binWidth - width of spatial bins
    
    binHeight - height of spatial bins
    
    numBinsX - number of bins in x direction for each descriptor
    
    numBinsY - number of bins in y direction for each descriptor
    
    numOriBins - number of orientation bins for each descriptor
  - DenseSIFT
```
public DenseSIFT(int stepX,
                 int stepY,
                 int binWidth,
                 int binHeight,
                 int numBinsX,
                 int numBinsY,
                 int numOriBins,
                 float gaussianWindowSize)
```
    Construct with the given configuration. The value threshold is set to 0.2.
    
    Parameters:
    
    stepX - step size in x direction
    
    stepY - step size in y direction
    
    binWidth - width of spatial bins
    
    binHeight - height of spatial bins
    
    numBinsX - number of bins in x direction for each descriptor
    
    numBinsY - number of bins in y direction for each descriptor
    
    numOriBins - number of orientation bins for each descriptor
    
    gaussianWindowSize - the size of the gaussian weighting window
  - DenseSIFT
```
public DenseSIFT(int stepX,
                 int stepY,
                 int binWidth,
                 int binHeight,
                 int numBinsX,
                 int numBinsY,
                 int numOriBins,
                 float gaussianWindowSize,
                 float valueThreshold)
```
    Construct with the given configuration.
    
    Parameters:
    
    stepX - step size in x direction
    
    stepY - step size in y direction
    
    binWidth - width of spatial bins
    
    binHeight - height of spatial bins
    
    numBinsX - number of bins in x direction for each descriptor
    
    numBinsY - number of bins in y direction for each descriptor
    
    numOriBins - number of orientation bins for each descriptor
    
    gaussianWindowSize - the size of the gaussian weighting window
    
    valueThreshold - the threshold for clipping features
- Method Detail
  - extractFeatures
```
protected void extractFeatures()
```
    Extract the DSIFT features
  - analyseImage
```
public void analyseImage(FImage image,
                         Rectangle bounds)
```
    Description copied from class: AbstractDenseSIFT
    
    Compute the dense sift descriptors inside the bounds rectangle of the given image.
    
    Specified by:
    
    analyseImage in class AbstractDenseSIFT<FImage>
    
    Parameters:
    
    image - the image
    
    bounds - the bounds rectangle
  - getFloatKeypoints
```
public LocalFeatureList<FloatDSIFTKeypoint> getFloatKeypoints()
```
    Description copied from class: AbstractDenseSIFT
    
    Get the SIFT descriptors from the previous call to AbstractDenseSIFT.analyseImage(Image) or AbstractDenseSIFT.analyseImage(Image, Rectangle) in the form of a list of local features with float vectors.
    
    Specified by:
    
    getFloatKeypoints in class AbstractDenseSIFT<FImage>
    
    Returns:
    
    a list of FloatDSIFTKeypoints.
  - getByteKeypoints
```
public LocalFeatureList<ByteDSIFTKeypoint> getByteKeypoints()
```
    Description copied from class: AbstractDenseSIFT
    
    Get the SIFT descriptors from the previous call to AbstractDenseSIFT.analyseImage(Image) or AbstractDenseSIFT.analyseImage(Image, Rectangle) in the form of a list of local features with byte vectors.
    
    Specified by:
    
    getByteKeypoints in class AbstractDenseSIFT<FImage>
    
    Returns:
    
    a list of ByteDSIFTKeypoints.
  - getFloatKeypoints
```
public LocalFeatureList<FloatDSIFTKeypoint> getFloatKeypoints(float energyThreshold)
```
    Description copied from class: AbstractDenseSIFT
    
    Get the SIFT descriptors from the previous call to AbstractDenseSIFT.analyseImage(Image) or AbstractDenseSIFT.analyseImage(Image, Rectangle) in the form of a list of local features with float vectors. Only the features with an energy above the given threshold will be returned.
    
    Specified by:
    
    getFloatKeypoints in class AbstractDenseSIFT<FImage>
    
    Parameters:
    
    energyThreshold - the threshold on the feature energy
    
    Returns:
    
    a list of FloatDSIFTKeypoints.
  - getByteKeypoints
```
public LocalFeatureList<ByteDSIFTKeypoint> getByteKeypoints(float energyThreshold)
```
    Description copied from class: AbstractDenseSIFT
    
    Get the SIFT descriptors from the previous call to AbstractDenseSIFT.analyseImage(Image) or AbstractDenseSIFT.analyseImage(Image, Rectangle) in the form of a list of local features with byte vectors. Only the features with an energy above the given threshold will be returned.
    
    Specified by:
    
    getByteKeypoints in class AbstractDenseSIFT<FImage>
    
    Parameters:
    
    energyThreshold - the threshold on the feature energy
    
    Returns:
    
    a list of ByteDSIFTKeypoints.
  - getDescriptors
```
public float[][] getDescriptors()
```
    Get the computed raw dense SIFT descriptors from the previous call to #analyseImage(FImage) or analyseImage(FImage, Rectangle) .
    
    Specified by:
    
    getDescriptors in class AbstractDenseSIFT<FImage>
    
    Returns:
    
    the descriptors.
  - clone
```
public DenseSIFT clone()
```
    Overrides:
    
    clone in class AbstractDenseSIFT<FImage>
  - setBinWidth
```
public void setBinWidth(int size)
```
    Description copied from class: AbstractDenseSIFT
    
    (Optional operation) Set the width of a single bin of the sampling window (in pixels). Sampling window width is this multiplied by AbstractDenseSIFT.getNumBinsX().
    
    Specified by:
    
    setBinWidth in class AbstractDenseSIFT<FImage>
    
    Parameters:
    
    size - size to set
  - setBinHeight
```
public void setBinHeight(int size)
```
    Description copied from class: AbstractDenseSIFT
    
    (Optional operation) Set the height of a single bin of the sampling window (in pixels). Sampling window height is this multiplied by AbstractDenseSIFT.getNumBinsY().
    
    Specified by:
    
    setBinHeight in class AbstractDenseSIFT<FImage>
    
    Parameters:
    
    size - size to set
  - getBinWidth
```
public int getBinWidth()
```
    Description copied from class: AbstractDenseSIFT
    
    (Optional operation) Get the width of a single bin of the sampling window (in pixels). Sampling window width is this multiplied by AbstractDenseSIFT.getNumBinsX().
    
    Specified by:
    
    getBinWidth in class AbstractDenseSIFT<FImage>
    
    Returns:
    
    the bin width
  - getBinHeight
```
public int getBinHeight()
```
    Description copied from class: AbstractDenseSIFT
    
    (Optional operation) Get the height of a single bin of the sampling window (in pixels). Sampling window height is this multiplied by AbstractDenseSIFT.getNumBinsY().
    
    Specified by:
    
    getBinHeight in class AbstractDenseSIFT<FImage>
    
    Returns:
    
    the bin height
  - getNumBinsX
```
public int getNumBinsX()
```
    Description copied from class: AbstractDenseSIFT
    
    Get the number of spatial bins in the X direction
    
    Specified by:
    
    getNumBinsX in class AbstractDenseSIFT<FImage>
    
    Returns:
    
    the number of bins in the x direction
  - getNumBinsY
```
public int getNumBinsY()
```
    Description copied from class: AbstractDenseSIFT
    
    Get the number of spatial bins in the Y direction
    
    Specified by:
    
    getNumBinsY in class AbstractDenseSIFT<FImage>
    
    Returns:
    
    the number of bins in the y direction
  - getNumOriBins
```
public int getNumOriBins()
```
    Description copied from class: AbstractDenseSIFT
    
    Get the number of orientation bins
    
    Specified by:
    
    getNumOriBins in class AbstractDenseSIFT<FImage>
    
    Returns:
    
    the number of orientation bins

Class DenseSIFT

Field Summary

Constructor Summary

Method Summary

Methods inherited from class org.openimaj.image.feature.dense.gradient.dsift.AbstractDenseSIFT

Methods inherited from class java.lang.Object

Field Detail

stepX

stepY

binWidth

binHeight

numBinsX

numBinsY

numOriBins

gaussianWindowSize

valueThreshold

data

descriptors

energies

Constructor Detail

DenseSIFT

DenseSIFT

DenseSIFT

DenseSIFT

DenseSIFT

Method Detail

extractFeatures

analyseImage

getFloatKeypoints

getByteKeypoints

getFloatKeypoints

getByteKeypoints

getDescriptors

clone

setBinWidth

setBinHeight

getBinWidth

getBinHeight

getNumBinsX

getNumBinsY

getNumOriBins