net.semanticmetadata.lire.indexers.tools

类 ParallelExtractor

java.lang.Object

net.semanticmetadata.lire.indexers.tools.ParallelExtractor

所有已实现的接口:

java.lang.Runnable

public class ParallelExtractor
extends java.lang.Object
implements java.lang.Runnable

The Extractor is a configurable class that extracts multiple features from multiple images and puts them into a data file. Main purpose is run multiple extractors at multiple machines and put the data files into one single index. Images are references relatively to the data file, so it should work fine for network file systems.

File format is specified as: (12(345)+('-1'))+ with 1-5 being ...

1. Length of the file name [4 bytes], an int n giving the number of bytes for the file name 2. File name, relative to the outfile [n bytes, see above] 3. Feature index [1 byte], see static members 4. Feature value length [4 bytes], an int k giving the number of bytes encoding the value 5. Feature value [k bytes, see above]

The file is sent through an GZIPOutputStream, so it's compressed in addition.

Note that the outfile has to be in a folder parent to all images!

// TODO: Change to LinkedBlockingQueue and Files.readAllBytes.

作者:

Mathias Lux, mathias@juggle.at, 08.03.13

字段概要

字段

限定符和类型	字段和说明
static java.lang.String[]	featureFieldNames
static java.lang.String[]	features

构造器概要

构造器

构造器和说明
ParallelExtractor()

方法概要

限定符和类型	方法和说明
void	addFeature(GlobalFeature feature) Adds a feature to the extractor chain.
int	getMaxSideLength()
static void	main(java.lang.String[] args)
void	run()
void	setFileList(java.io.File fileList) Sets the file list for processing.
void	setMaxSideLength(int maxSideLength)
static void	setNumberOfThreads(int numberOfThreads) Sets the number of consumer threads that are employed for extraction
void	setOutFile(java.io.File outFile) Sets the outfile.

从类继承的方法 java.lang.Object

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

字段详细资料

features

public static final java.lang.String[] features

featureFieldNames

public static final java.lang.String[] featureFieldNames

构造器详细资料

ParallelExtractor

public ParallelExtractor()

方法详细资料

setNumberOfThreads

public static void setNumberOfThreads(int numberOfThreads)

Sets the number of consumer threads that are employed for extraction

参数:

numberOfThreads -

main

public static void main(java.lang.String[] args)
                 throws java.io.IOException

抛出:

java.io.IOException

addFeature

public void addFeature(GlobalFeature feature)

Adds a feature to the extractor chain. All those features are extracted from images.

参数:

feature -

setFileList

public void setFileList(java.io.File fileList)

Sets the file list for processing. One image file per line is fine.

参数:

fileList -

setOutFile

public void setOutFile(java.io.File outFile)

Sets the outfile. The outfile has to be in a folder parent to all input images.

参数:

outFile -

getMaxSideLength

public int getMaxSideLength()

setMaxSideLength

public void setMaxSideLength(int maxSideLength)

run

public void run()

指定者:

run 在接口中 java.lang.Runnable