net.semanticmetadata.lire.builders

类 GlobalDocumentBuilder

java.lang.Object

net.semanticmetadata.lire.builders.GlobalDocumentBuilder

所有已实现的接口:

DocumentBuilder

public class GlobalDocumentBuilder
extends java.lang.Object
implements DocumentBuilder

This class creates Lucene Documents from images using one or multiple Global Features. Can also be used only for extraction.

作者:

Nektarios Anagnostopoulos, nek.anag@gmail.com (c) 2015 by Nektarios Anagnostopoulos, Mathias Lux, mathias@juggle.at

嵌套类概要

嵌套类

限定符和类型	类和说明
static class	GlobalDocumentBuilder.HashingMode

字段概要

从接口继承的字段 net.semanticmetadata.lire.builders.DocumentBuilder

FIELD_NAME_ACCID, FIELD_NAME_AUTOCOLORCORRELOGRAM, FIELD_NAME_BINARY_PATTERNS_PYRAMID, FIELD_NAME_CEDD, FIELD_NAME_COLORHISTOGRAM, FIELD_NAME_COLORLAYOUT, FIELD_NAME_CVOPPSIFT, FIELD_NAME_CVSIFT, FIELD_NAME_CVSURF, FIELD_NAME_EDGEHISTOGRAM, FIELD_NAME_FCTH, FIELD_NAME_GABOR, FIELD_NAME_IDENTIFIER, FIELD_NAME_JCD, FIELD_NAME_JOINT_HISTOGRAM, FIELD_NAME_JPEGCOEFFS, FIELD_NAME_LOCAL_BINARY_PATTERNS, FIELD_NAME_LOCAL_BINARY_PATTERNS_AND_OPPONENT, FIELD_NAME_LUMINANCE_LAYOUT, FIELD_NAME_OPPONENT_HISTOGRAM, FIELD_NAME_PHOG, FIELD_NAME_Rank_Opponent, FIELD_NAME_ROTATION_INVARIANT_LOCAL_BINARY_PATTERNS, FIELD_NAME_SCALABLECOLOR, FIELD_NAME_SELF_SIMILARITIES, FIELD_NAME_SELF_SIMILARITIES_ORIG, FIELD_NAME_SIFT, FIELD_NAME_SIMPLE, FIELD_NAME_SURF, FIELD_NAME_TAMURA, HASH_FIELD_SUFFIX, MAX_IMAGE_DIMENSION, NUM_OF_THREADS

构造器概要

构造器

构造器和说明
GlobalDocumentBuilder()
GlobalDocumentBuilder(boolean hashing)
GlobalDocumentBuilder(boolean hashing, GlobalDocumentBuilder.HashingMode hashingMode) Creates a GlobalDocumentBuilder with the specific hashing mode.
GlobalDocumentBuilder(boolean hashing, GlobalDocumentBuilder.HashingMode hashingMode, boolean useDocValues) Creates a GlobalDocumentBuilder with the specific hashing mode.
GlobalDocumentBuilder(java.lang.Class<? extends GlobalFeature> globalFeatureClass)
GlobalDocumentBuilder(java.lang.Class<? extends GlobalFeature> globalFeatureClass, boolean hashing)
GlobalDocumentBuilder(java.lang.Class<? extends GlobalFeature> globalFeatureClass, boolean hashing, boolean useDocValues) Use DocValues instead of TextField implementations for storing the feature vector.
GlobalDocumentBuilder(ExtractorItem extractorItem)

方法概要

限定符和类型	方法和说明
void	addExtractor(java.lang.Class<? extends GlobalFeature> globalFeatureClass) Can be used to add global extractors.
void	addExtractor(ExtractorItem extractorItem) Can be used to add global extractors.
org.apache.lucene.document.Field[]	createDescriptorFields(java.awt.image.BufferedImage image) Creates the feature fields for a Lucene Document without creating the document itself.
org.apache.lucene.document.Document	createDocument(java.awt.image.BufferedImage image, java.lang.String identifier) Creates a new Lucene document from a BufferedImage.
GlobalFeature	extractGlobalFeature(java.awt.image.BufferedImage image, GlobalFeature globalFeature) Images are resized so as not to exceed the DocumentBuilder.MAX_IMAGE_DIMENSION, after that the feature is extracted using the given globalFeature.

从类继承的方法 java.lang.Object

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

构造器详细资料

GlobalDocumentBuilder

public GlobalDocumentBuilder()

GlobalDocumentBuilder

public GlobalDocumentBuilder(boolean hashing)

GlobalDocumentBuilder

public GlobalDocumentBuilder(boolean hashing,
                             GlobalDocumentBuilder.HashingMode hashingMode)

Creates a GlobalDocumentBuilder with the specific hashing mode. Please note that you have to take care of the initilization of the hashing subsystem yourself.

参数:

hashing - true if you want hashing to be applied.

hashingMode - the actual mode, eg. BitSampling or MetricSpaces.

GlobalDocumentBuilder

public GlobalDocumentBuilder(boolean hashing,
                             GlobalDocumentBuilder.HashingMode hashingMode,
                             boolean useDocValues)

Creates a GlobalDocumentBuilder with the specific hashing mode. Please note that you have to take care of the initilization of the hashing subsystem yourself. Optionally use DocValues instead of TextField implementations for storing the feature vector. Note that this cannot be read by ordinary linear searchers, but must be implemented in a different way.

参数:

hashing - true if you want hashing to be applied.

hashingMode - the actual mode, eg. BitSampling or MetricSpaces.

useDocValues - set to true if you want to use DocValues instead of Lucene fields.

GlobalDocumentBuilder

public GlobalDocumentBuilder(java.lang.Class<? extends GlobalFeature> globalFeatureClass)

GlobalDocumentBuilder

public GlobalDocumentBuilder(java.lang.Class<? extends GlobalFeature> globalFeatureClass,
                             boolean hashing)

GlobalDocumentBuilder

public GlobalDocumentBuilder(java.lang.Class<? extends GlobalFeature> globalFeatureClass,
                             boolean hashing,
                             boolean useDocValues)

Use DocValues instead of TextField implementations for storing the feature vector. Note that this cannot be read by ordinary linear searchers, but must be implmented in a different way.

参数:

globalFeatureClass -

hashing - set to true if hashing should be performed.

useDocValues - set to true if you want to use DocValues instead of Lucene fields.

GlobalDocumentBuilder

public GlobalDocumentBuilder(ExtractorItem extractorItem)

方法详细资料

addExtractor

public void addExtractor(java.lang.Class<? extends GlobalFeature> globalFeatureClass)

Can be used to add global extractors.

参数:

globalFeatureClass -

addExtractor

public void addExtractor(ExtractorItem extractorItem)

Can be used to add global extractors.

参数:

extractorItem -

extractGlobalFeature

public GlobalFeature extractGlobalFeature(java.awt.image.BufferedImage image,
                                          GlobalFeature globalFeature)

Images are resized so as not to exceed the DocumentBuilder.MAX_IMAGE_DIMENSION, after that the feature is extracted using the given globalFeature.

参数:

image - is the image

globalFeature - selected global feature

返回:

the input globalFeature

createDescriptorFields

public org.apache.lucene.document.Field[] createDescriptorFields(java.awt.image.BufferedImage image)

从接口复制的说明: DocumentBuilder

Creates the feature fields for a Lucene Document without creating the document itself.

指定者:

createDescriptorFields 在接口中 DocumentBuilder

参数:

image - the image to analyze.

返回:

Lucene Fields.

createDocument

public org.apache.lucene.document.Document createDocument(java.awt.image.BufferedImage image,
                                                          java.lang.String identifier)

从接口复制的说明: DocumentBuilder

Creates a new Lucene document from a BufferedImage. The identifier can be used like an id (e.g. the file hashFunctionsFileName or the url of the image)

指定者:

createDocument 在接口中 DocumentBuilder

参数:

image - the image to index. Cannot be NULL.

identifier - an id for the image, for instance the filename or a URL. Can be NULL.

返回:

a Lucene Document.