org.openimaj.image.objectdetection.haar

Class HaarFeature

java.lang.Object

org.openimaj.image.objectdetection.haar.HaarFeature

@Reference(type=Inproceedings,
           author={"Viola, P.","Jones, M."},
           title="Rapid object detection using a boosted cascade of simple features",
           year="2001",
           booktitle="Computer Vision and Pattern Recognition, 2001. CVPR 2001. Proceedings of the 2001 IEEE Computer Society Conference on",
           pages={" I","511 "," I","518 vol.1"},
           number="",
           volume="1",
           customData={"keywords"," AdaBoost; background regions; boosted simple feature cascade; classifiers; face detection; image processing; image representation; integral image; machine learning; object specific focus-of-attention mechanism; rapid object detection; real-time applications; statistical guarantees; visual object detection; feature extraction; image classification; image representation; learning (artificial intelligence); object detection;","doi","10.1109/CVPR.2001.990517","ISSN","1063-6919 "})
public abstract class HaarFeature
extends Object

Class describing a Haar-like feature. The features are typically built from two or three overlapping rectangles, and can represent edges, lines and centre-surround features.

The response of applying the feature to a specific point on an image (with a specific scaling) can be efficiently calculated using summed area tables.

Internally this implementation caches a scaled version of each rectangle for a given detection scale.

Author:

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

Field Summary

Fields

Modifier and Type	Field and Description
protected WeightedRectangle[]	cachedRects
WeightedRectangle[]	rects The rectangles that make up this feature.

Method Summary

Modifier and Type	Method and Description
abstract float	computeResponse(SummedSqTiltAreaTable sat, int x, int y) Compute the response of this feature at the given location.
static HaarFeature	create(boolean tilted, int x0, int y0, int w0, int h0, float wt0, int x1, int y1, int w1, int h1, float wt1) Construct a feature with the given parameters.
static HaarFeature	create(boolean tilted, int x0, int y0, int w0, int h0, float wt0, int x1, int y1, int w1, int h1, float wt1, int x2, int y2, int w2, int h2, float wt2) Construct a feature with the given parameters.
static HaarFeature	create(List<WeightedRectangle> rectList, boolean tilted) Create a feature from the given data.
void	setScale(float scale, float invArea) Set the current detection scale, setting up the internal caches appropriately.

Methods inherited from class java.lang.Object

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

Field Detail

rects

public WeightedRectangle[] rects

The rectangles that make up this feature.

cachedRects

protected WeightedRectangle[] cachedRects

Method Detail

setScale

public final void setScale(float scale,
                           float invArea)

Set the current detection scale, setting up the internal caches appropriately.

Parameters:

scale - the scale

invArea - the inverse of the detector area

computeResponse

public abstract float computeResponse(SummedSqTiltAreaTable sat,
                                      int x,
                                      int y)

Compute the response of this feature at the given location. The scale of the feature must have previously been set through a call to setScale(float, float) (this is only required once per scale).

Parameters:

sat - the summed area table(s). If there are tilted features, then this must include the tilted SAT.

x - the x-ordinate for the window being tested

y - the y-ordinate for the window being tested

Returns:

the response to the feature

create

public static HaarFeature create(List<WeightedRectangle> rectList,
                                 boolean tilted)

Create a feature from the given data. The specific type of feature created depends on whether or not the feature is tilted.

Parameters:

rectList - the rectangles defining the feature

tilted - is the feature tilted?

Returns:

the new HaarFeature object.

create

public static HaarFeature create(boolean tilted,
                                 int x0,
                                 int y0,
                                 int w0,
                                 int h0,
                                 float wt0,
                                 int x1,
                                 int y1,
                                 int w1,
                                 int h1,
                                 float wt1)

Construct a feature with the given parameters.

Parameters:

tilted - is the feature tilted?

x0 - x-ordinate of top-left of first rectangle

y0 - y-ordinate of top-left of first rectangle

w0 - width of first rectangle

h0 - height of first rectangle

wt0 - weight of first rectangle

x1 - x-ordinate of top-left of second rectangle

y1 - y-ordinate of top-left of second rectangle

w1 - width of second rectangle

h1 - height of second rectangle

wt1 - weight of second rectangle

Returns:

the feature

create

public static HaarFeature create(boolean tilted,
                                 int x0,
                                 int y0,
                                 int w0,
                                 int h0,
                                 float wt0,
                                 int x1,
                                 int y1,
                                 int w1,
                                 int h1,
                                 float wt1,
                                 int x2,
                                 int y2,
                                 int w2,
                                 int h2,
                                 float wt2)

Construct a feature with the given parameters.

Parameters:

tilted - is the feature tilted?

x0 - x-ordinate of top-left of first rectangle

y0 - y-ordinate of top-left of first rectangle

w0 - width of first rectangle

h0 - height of first rectangle

wt0 - weight of first rectangle

x1 - x-ordinate of top-left of second rectangle

y1 - y-ordinate of top-left of second rectangle

w1 - width of second rectangle

h1 - height of second rectangle

wt1 - weight of second rectangle

x2 - x-ordinate of top-left of third rectangle

y2 - y-ordinate of top-left of third rectangle

w2 - width of third rectangle

h2 - height of third rectangle

wt2 - weight of third rectangle

Returns:

the feature