org.openimaj.image.model

Interface ImageClassificationModel<T extends org.openimaj.image.Image<?,T>>

Type Parameters:

T - the type of image that the model can be applied to

All Superinterfaces:

Cloneable, org.openimaj.math.model.EstimatableModel<T,org.openimaj.image.FImage>, org.openimaj.math.model.Model<T,org.openimaj.image.FImage>, Serializable

All Known Implementing Classes:

HistogramPatchModel, HistogramPixelModel, MBFPatchClassificationModel, MBFPixelClassificationModel, OrthotopePixelModel, PatchClassificationModel, PixelClassificationModel, SingleGaussianPixelModel

public interface ImageClassificationModel<T extends org.openimaj.image.Image<?,T>>
extends org.openimaj.math.model.EstimatableModel<T,org.openimaj.image.FImage>, Serializable

An ImageClassificationModel is a EstimatableModel constructed between an generic image and a probability map in the form of an FImage. Potential uses for such a model are for the prediction of certain classes of pixels in an image. For example, a model could be constructed that predicted skin-tones in an image based on hue and saturation values of pixels. With such a model, a colour image could be presented, and a probability map would be returned.

Author:

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

Method Summary

Modifier and Type	Method and Description
org.openimaj.image.FImage	classifyImage(T im) Classify the given image and return the corresponding probability map
ImageClassificationModel<T>	clone()
void	learnModel(T... images) Learn the model from the given MBFImages.

Methods inherited from interface org.openimaj.math.model.EstimatableModel

estimate, numItemsToEstimate

Methods inherited from interface org.openimaj.math.model.Model

predict

Method Detail

learnModel

void learnModel(T... images)

Learn the model from the given MBFImages.

Parameters:

images - the images to learn from

classifyImage

org.openimaj.image.FImage classifyImage(T im)

Classify the given image and return the corresponding probability map

Parameters:

im - the image to classify

Returns:

the probability map

clone

ImageClassificationModel<T> clone()

Specified by:

clone in interface org.openimaj.math.model.EstimatableModel<T extends org.openimaj.image.Image<?,T>,org.openimaj.image.FImage>