Package boofcv.abst.scene

Interface SceneRecognition<T extends ImageBase<T>>

All Superinterfaces:

VerbosePrint

All Known Implementing Classes:

WrapFeatureToSceneRecognition

public interface SceneRecognition<T extends ImageBase<T>>
extends VerbosePrint

Implementations of this interface seek to solve the problem of "have I seen this scene before, but from the same or different perspective? If so find those images". For example, this should be used if you want to find images of a mountain taken from a different angle or zoom. Implementations may or may not apply geometric constraints to ensure that it is the same scene. Usage Example:

1. Learn a model from a set of images by calling learnModel(java.util.Iterator<T>)
2. Add images for later retrieval by calling addImage(java.lang.String, T)
3. Call query(T, boofcv.misc.BoofLambdas.Filter<java.lang.String>, int, org.ddogleg.struct.DogArray<boofcv.abst.scene.SceneRecognition.Match>) to find the set of N images which are most similar to the passed in image

You can also save models and your image database by using functions in RecognitionIO.

Nested Class Summary

Nested Classes

Modifier and Type	Interface	Description
static class	SceneRecognition.Match	References a match in the database to the query image

Method Summary

Modifier and Type	Method	Description
void	addImage(String id, T image)	Adds a new image to the database
void	clearDatabase()	Removes all images from the database.
List<String>	getImageIds(@Nullable List<String> storage)	Returns a list of image IDs in the database
ImageType<T>	getImageType()	The image data type which can be processed
void	learnModel(Iterator<T> images)	Learns a model by finding the most distinctive features in the provided set of images.
boolean	query(T queryImage, BoofLambdas.Filter<String> filter, int limit, DogArray<SceneRecognition.Match> matches)	Finds the best matches in the database to the query image.

Methods inherited from interface org.ddogleg.struct.VerbosePrint

setVerbose

Method Details

learnModel

void learnModel(Iterator<T> images)

Learns a model by finding the most distinctive features in the provided set of images. Images are not added to the database.

clearDatabase

void clearDatabase()

Removes all images from the database. The model is not modified

addImage

void addImage(String id,
 T image)

Adds a new image to the database

Parameters:

id - The unique ID for this image

image - The image

query

boolean query(T queryImage,
 @Nullable
 BoofLambdas.Filter<String> filter,
 int limit,
 DogArray<SceneRecognition.Match> matches)

Finds the best matches in the database to the query image.

Parameters:

queryImage - (Input) image being processed

filter - (Input) Used to filter results so that known matches don't pollute the results.

limit - (Input) The maximum number of results it will return. If ≤ 0 then all matches are returned.

matches - (Output) Set of matches found in best first order. List is always cleared

Returns:

true if at least one valid match was found or false if no valid matches could be found. If false that means matches is empty. This is strictly a convenience.

getImageIds

List<String> getImageIds(@Nullable
 @Nullable List<String> storage)

Returns a list of image IDs in the database

Parameters:

storage - (Optional) Storage for the list of images. If null a new instance is created

Returns:

List of all the image IDs.

getImageType

ImageType<T> getImageType()

The image data type which can be processed