org.openimaj.image.saliency

Class LuoTangSubjectRegion

java.lang.Object

org.openimaj.image.saliency.LuoTangSubjectRegion

All Implemented Interfaces:

ImageAnalyser<FImage>, SaliencyMapGenerator<FImage>

@Reference(type=Inproceedings,
           author={"Luo, Yiwen","Tang, Xiaoou"},
           title="Photo and Video Quality Evaluation: Focusing on the Subject",
           year="2008",
           booktitle="Proceedings of the 10th European Conference on Computer Vision: Part III",
           pages={"386","","399"},
           url="http://dx.doi.org/10.1007/978-3-540-88690-7_29",
           publisher="Springer-Verlag",
           series="ECCV \'08",
           customData={"isbn","978-3-540-88689-1","location","Marseille, France","numpages","14","doi","10.1007/978-3-540-88690-7_29","acmid","1478204","address","Berlin, Heidelberg"})
public class LuoTangSubjectRegion
extends Object
implements SaliencyMapGenerator<FImage>

Extract the subject region of an image based on the part that is least blurred (most in-focus).

Algorithm based on: Yiwen Luo and Xiaoou Tang. 2008. Photo and Video Quality Evaluation: Focusing on the Subject. In Proceedings of the 10th European Conference on Computer Vision: Part III (ECCV '08), David Forsyth, Philip Torr, and Andrew Zisserman (Eds.). Springer-Verlag, Berlin, Heidelberg, 386-399. DOI=10.1007/978-3-540-88690-7_29 http://dx.doi.org/10.1007/978-3-540-88690-7_29

Note that this is not scale invariant - you will get different results with different sized images...

Author:

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

Constructor Summary

Constructors

Constructor and Description
LuoTangSubjectRegion() Construct with default values for the DepthOfFieldEstimator and an alpha parameter of 0.9.
LuoTangSubjectRegion(float alpha, int maxKernelSize, int kernelSizeStep, int nbins, int windowSize) Construct with the given parameters.

Method Summary

Modifier and Type	Method and Description
void	analyseImage(FImage image) Analyse an image.
Rectangle	calculateROI()
FImage	getROIMap()
FImage	getSaliencyMap() Get the generated saliency map

Methods inherited from class java.lang.Object

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

Constructor Detail

LuoTangSubjectRegion

public LuoTangSubjectRegion()

Construct with default values for the DepthOfFieldEstimator and an alpha parameter of 0.9.

LuoTangSubjectRegion

public LuoTangSubjectRegion(float alpha,
                            int maxKernelSize,
                            int kernelSizeStep,
                            int nbins,
                            int windowSize)

Construct with the given parameters.

Parameters:

alpha - the alpha value.

maxKernelSize - Maximum kernel size for the DepthOfFieldEstimator.

kernelSizeStep - Kernel step size for the DepthOfFieldEstimator.

nbins - Number of bins for the DepthOfFieldEstimator.

windowSize - window size for the DepthOfFieldEstimator.

Method Detail

analyseImage

public void analyseImage(FImage image)

Description copied from interface: ImageAnalyser

Analyse an image.

Specified by:

analyseImage in interface ImageAnalyser<FImage>

Parameters:

image - The image to process in place.

calculateROI

public Rectangle calculateROI()

Returns:

the estimated rectangular region of interest

getSaliencyMap

public FImage getSaliencyMap()

Description copied from interface: SaliencyMapGenerator

Get the generated saliency map

Specified by:

getSaliencyMap in interface SaliencyMapGenerator<FImage>

Returns:

the saliency map

getROIMap

public FImage getROIMap()

Returns:

a mask image showing the region of interest.