Java LuceneUtils类代码示例

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
/**
 * Index a picture
 * @param source
 * @param picture_id
 * @param conf
 * @throws IOException
 */
public static void index(byte[] source, UUID picture_id, IndexWriterConfig conf) throws IOException
{
    ByteArrayInputStream in = new ByteArrayInputStream(source);
    BufferedImage image = ImageIO.read(in);

    // Creating an Lucene IndexWriter
    log.debug("Is Lucene configured? " + (conf == null));
    if(conf == null) {
        conf = new IndexWriterConfig(LuceneUtils.LUCENE_VERSION, new WhitespaceAnalyzer(LuceneUtils.LUCENE_VERSION));
        conf.setOpenMode(IndexWriterConfig.OpenMode.CREATE_OR_APPEND);
    }

    luceneIndexer(image, picture_id, FeatureEnumerate.AutoColorCorrelogram.getText(), DocumentBuilderFactory.getAutoColorCorrelogramDocumentBuilder(), conf);
    luceneIndexer(image, picture_id, FeatureEnumerate.CEDD.getText(), DocumentBuilderFactory.getCEDDDocumentBuilder(), conf);
    luceneIndexer(image, picture_id, FeatureEnumerate.ColorLayout.getText(), DocumentBuilderFactory.getColorLayoutBuilder(), conf);
    luceneIndexer(image, picture_id, FeatureEnumerate.EdgeHistogram.getText(), DocumentBuilderFactory.getEdgeHistogramBuilder(), conf);
    luceneIndexer(image, picture_id, FeatureEnumerate.ColorHistogram.getText(), DocumentBuilderFactory.getColorHistogramDocumentBuilder(), conf);
    luceneIndexer(image, picture_id, FeatureEnumerate.PHOG.getText(), DocumentBuilderFactory.getPHOGDocumentBuilder(), conf);

}
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
private static void deleteFromFeature(UUID pictureId, Term term, String prefix, IndexWriterConfig conf) throws IOException {

        File file = getPath(prefix);

        // Creating an Lucene IndexWriter
        log.debug("Is Lucene configured: " + (conf == null));
        if(conf == null) {
            conf = new IndexWriterConfig(LuceneUtils.LUCENE_VERSION, new WhitespaceAnalyzer(LuceneUtils.LUCENE_VERSION));
            conf.setOpenMode(IndexWriterConfig.OpenMode.CREATE_OR_APPEND);
        }
        IndexWriter iw = new IndexWriter(FSDirectory.open(file), conf);

        iw.deleteDocuments(term);

        iw.close();
    }
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
public void run() {
        // do it ...
        try {
//            IndexWriter indexWriter = LuceneUtils.createIndexWriter(indexPath, overwriteIndex, LuceneUtils.AnalyzerType.WhitespaceAnalyzer);
            IndexWriterConfig config = new IndexWriterConfig(LuceneUtils.LUCENE_VERSION, new WhitespaceAnalyzer(LuceneUtils.LUCENE_VERSION));
            config.setOpenMode(IndexWriterConfig.OpenMode.CREATE_OR_APPEND);
            config.setCodec(new LireCustomCodec());
            IndexWriter indexWriter = new IndexWriter(FSDirectory.open(new File(indexPath)), config);
            for (Iterator<File> iterator = inputFiles.iterator(); iterator.hasNext(); ) {
                File inputFile = iterator.next();
                if (verbose) System.out.println("Processing " + inputFile.getPath() + ".");
                readFile(indexWriter, inputFile);
                if (verbose) System.out.println("Indexing finished.");
            }
            indexWriter.commit();
            indexWriter.close();
        } catch (Exception e) {
            e.printStackTrace();
        }
    }
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
private void indexFiles(ArrayList<String> images, DocumentBuilder builder, String indexPath) throws IOException {
//        System.out.println(">> Indexing " + images.size() + " files.");
//        DocumentBuilder builder = DocumentBuilderFactory.getExtensiveDocumentBuilder();
//        DocumentBuilder builder = DocumentBuilderFactory.getFastDocumentBuilder();
        IndexWriter iw = LuceneUtils.createIndexWriter(indexPath, true);
        int count = 0;
        long time = System.currentTimeMillis();
        for (String identifier : images) {
            Document doc = builder.createDocument(new FileInputStream(identifier), identifier);
            iw.addDocument(doc);
            count++;
            if (count % 100 == 0) System.out.println(count + " files indexed.");
//            if (count == 200) break;
        }
        long timeTaken = (System.currentTimeMillis() - time);
        float sec = ((float) timeTaken) / 1000f;

        System.out.println(sec + " seconds taken, " + (timeTaken / count) + " ms per image.");
        iw.commit();
        iw.close();
    }
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
private void indexFiles(ArrayList<String> images, DocumentBuilder builder, String indexPath) throws IOException {
        System.out.println(">> Indexing " + images.size() + " files.");
        IndexWriter iw = LuceneUtils.createIndexWriter(indexPath, true);
        int count = 0;
        long time = System.currentTimeMillis();
        for (String identifier : images) {
            Document doc = builder.createDocument(new FileInputStream(identifier), identifier);
            iw.addDocument(doc);
            count++;
            if (count % 100 == 0) System.out.println(count + " files indexed.");
//            if (count == 200) break;
        }
        long timeTaken = (System.currentTimeMillis() - time);
        float sec = ((float) timeTaken) / 1000f;

        System.out.println(sec + " seconds taken, " + (timeTaken / count) + " ms per image.");
        iw.close();
    }
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
public void testIndexing() throws IOException {
    ChainedDocumentBuilder b = new ChainedDocumentBuilder();
    b.addBuilder(new SiftDocumentBuilder());
    // b.addBuilder(DocumentBuilderFactory.getCEDDDocumentBuilder());

    ArrayList<String> files = net.semanticmetadata.lire.utils.FileUtils.getAllImages(new File("W:\\MultimediaShare\\image_datasets\\ukbench-nister\\full"), true);

    System.out.println("files.size() = " + files.size());

    IndexWriter writer = LuceneUtils.createIndexWriter("nisterindex_sift", true);
    int count = 0;
    long ms = System.currentTimeMillis();
    for (Iterator<String> iterator = files.iterator(); iterator.hasNext(); ) {
        String next = iterator.next();
        Document d = b.createDocument(ImageIO.read(new File(next)), next);
        writer.addDocument(d);
        count++;
        if (count % 100 == 0) {
            float time = (float) (System.currentTimeMillis() - ms);
            System.out.println("Finished " + count + " images, " + (((float) count) / 10200f) * 100 + "%. " + (time / (float) count) + " ms per image.");
        }
    }
    writer.close();
}
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
private void indexFiles(String dir, String index, int featureIndex, boolean createNewIndex) throws IOException {
        ArrayList<String> images = FileUtils.getAllImages(new File(dir), true);
        IndexWriter iw = LuceneUtils.createIndexWriter(index, createNewIndex, LuceneUtils.AnalyzerType.WhitespaceAnalyzer);
        // select one feature for the large index:
        int count = 0;
        long ms = System.currentTimeMillis();
        DocumentBuilder builder = new ChainedDocumentBuilder();
        ((ChainedDocumentBuilder) builder).addBuilder(builders[featureIndex]);
//        ((ChainedDocumentBuilder) builder).addBuilder(builders[0]);
        for (Iterator<String> iterator = images.iterator(); iterator.hasNext(); ) {
            count++;
            if (count > 100 && count % 5000 == 0) {
                System.out.println(count + " files indexed. " + (System.currentTimeMillis() - ms) / (count) + " ms per file");
            }
            String file = iterator.next();
            try {
                iw.addDocument(builder.createDocument(new FileInputStream(file), file));
            } catch (Exception e) {
                System.err.println("Error: " + e.getMessage());
            }
        }
        iw.close();
    }
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
public void testIndexMissingFiles() throws IOException {
        // first delete some of the existing ones ...
        System.out.println("Deleting visual words from docs ...");
        IndexReader ir = DirectoryReader.open(FSDirectory.open(new File(index)));
        IndexWriter iw = LuceneUtils.createIndexWriter(index, false);
        int maxDocs = ir.maxDoc();
        for (int i = 0; i < maxDocs / 10; i++) {
            Document d = ir.document(i);
            d.removeFields(DocumentBuilder.FIELD_NAME_SURF_VISUAL_WORDS);
            d.removeFields(DocumentBuilder.FIELD_NAME_SURF_LOCAL_FEATURE_HISTOGRAM);
//            d.removeFields(DocumentBuilder.FIELD_NAME_SURF);
            iw.updateDocument(new Term(DocumentBuilder.FIELD_NAME_IDENTIFIER, d.getValues(DocumentBuilder.FIELD_NAME_IDENTIFIER)[0]), d);
        }
        System.out.println("# of deleted docs:  " + maxDocs / 10);
        System.out.println("Optimizing and closing ...");
        iw.close();
        ir.close();
        System.out.println("Creating new visual words ...");
        SurfFeatureHistogramBuilder surfFeatureHistogramBuilder = new SurfFeatureHistogramBuilder(DirectoryReader.open(FSDirectory.open(new File(index))), numSamples, clusters);
        surfFeatureHistogramBuilder.indexMissing();
        System.out.println("Finished.");
    }
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
public void testCreateCEDDIndex() throws IOException {
    ArrayList<String> images = FileUtils.getAllImages(new File("E:/temp/flickrphotos/flickrphotos"), true);

    ChainedDocumentBuilder builder = new ChainedDocumentBuilder();
    builder.addBuilder(DocumentBuilderFactory.getCEDDDocumentBuilder());
    IndexWriter iw = LuceneUtils.createIndexWriter(indexPath + "-cedd", true);
    int count = 0;
    long ms = System.currentTimeMillis();
    for (String identifier : images) {
        try {
            Document doc = builder.createDocument(new FileInputStream(identifier), identifier);
            iw.addDocument(doc);
        } catch (Exception e) {
            System.err.print("\n ;-( ");//e.printStackTrace();
        }
        count++;
        if (count % 1000 == 0) System.out.println((100 * count) / images.size() + "% ");
    }
    System.out.println("Time taken: " + ((System.currentTimeMillis() - ms) / testFiles.length) + " ms");
    iw.commit();
    iw.close();
}
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
private void indexFiles(String prefix, ArrayList<String> images, DocumentBuilder builder, String indexPath) throws IOException {
    System.out.println(">> Indexing " + images.size() + " files.");
    IndexWriter iw = LuceneUtils.createIndexWriter(indexPath, true);
    int count = 0;
    long time = System.currentTimeMillis();
    for (String identifier : images) {
        Document doc = builder.createDocument(new FileInputStream(identifier), identifier);
        iw.addDocument(doc);
        count++;
        if (count % 100 == 0) System.out.print((100 * count) / images.size() + "% ");
    }
    long timeTaken = (System.currentTimeMillis() - time);
    float sec = ((float) timeTaken) / 1000f;
    System.out.println("");
    System.out.println(prefix + sec + " seconds taken, " + (timeTaken / count) + " ms per image.");
    // iw.optimize();
    iw.close();
}
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
public void testExtendedIndexMSER() throws IOException {
    MSERDocumentBuilder builder = new MSERDocumentBuilder();
    IndexWriterConfig conf = new IndexWriterConfig(LuceneUtils.LUCENE_VERSION,
            new WhitespaceAnalyzer(LuceneUtils.LUCENE_VERSION));
    IndexWriter iw = new IndexWriter(FSDirectory.open(indexPath), conf);
    long ms = System.currentTimeMillis();
    int count = 0;
    ArrayList<File> files = FileUtils.getAllImageFiles(new File("D:\\DataSets\\WIPO\\CA\\sample"), true);
    for (Iterator<File> i = files.iterator(); i.hasNext(); ) {
        File imgFile = i.next();
        BufferedImage img = ImageIO.read(imgFile);
        if (Math.max(img.getWidth(), img.getHeight()) < 800) {
            // scale image ...
            img = ImageUtils.scaleImage(img, 800);
        }
        iw.addDocument(builder.createDocument(img, imgFile.getPath()));
        count++;
        if (count > 2 && count % 25 == 0) {
            System.out.println(count + " files indexed. " + (System.currentTimeMillis() - ms) / (count) + " ms per file");
        }

    }
    iw.close();
}
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
public void testSiftIndexing() throws IOException {
        ArrayList<String> images = FileUtils.getAllImages(new File(testExtensive), true);
        ChainedDocumentBuilder db = new ChainedDocumentBuilder();
        db.addBuilder(new SiftDocumentBuilder());
        db.addBuilder(DocumentBuilderFactory.getCEDDDocumentBuilder());
        IndexWriter iw = LuceneUtils.createIndexWriter("sift-idx", true);

        for (int i = 0; i < images.size(); i++) {
//            int sampleQuery = sampleQueries[i];
//            String s = testExtensive + "/" + sampleQuery + ".jpg";
            iw.addDocument(db.createDocument(new FileInputStream(images.get(i)), images.get(i)));
            if (i % 100 == 99) System.out.print(".");
            if (i % 1000 == 999) System.out.print(" ~ " + i + " files indexed\n");
            if (i > 1000) break;
        }
        System.out.println("");
        iw.close();
    }
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
public void testSurfIndexing() throws IOException {
        ArrayList<String> images = FileUtils.getAllImages(new File(testExtensive), true);
        ChainedDocumentBuilder db = new ChainedDocumentBuilder();
        db.addBuilder(new SurfDocumentBuilder());
        IndexWriter iw = LuceneUtils.createIndexWriter("sift-idx", true);
        for (int i = 0; i < images.size(); i++) {
//            int sampleQuery = sampleQueries[i];
//            String s = testExtensive + "/" + sampleQuery + ".jpg";
            iw.addDocument(db.createDocument(new FileInputStream(images.get(i)), images.get(i)));
            if (i % 100 == 99) System.out.print(".");
            if (i % 1000 == 999) System.out.print(" ~ " + i + " files indexed\n");
            if (i > 1000) break;
        }
        System.out.println("");
        iw.close();
    }
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
@SuppressWarnings("unused")
private Document indexFiles() throws IOException {
       System.out.println("---< indexing >-------------------------");
       int count = 0;
       DocumentBuilder builder = getDocumentBuilder();
       ArrayList<String> allImages = FileUtils.getAllImages(new File("wang-1000"), true);
       IndexWriter iw = LuceneUtils.createIndexWriter(indexPath, true);
       Document document = null;
       for (Iterator<String> iterator = allImages.iterator(); iterator.hasNext(); ) {
           String filename = iterator.next();
           BufferedImage image = ImageIO.read(new FileInputStream(filename));
           document = builder.createDocument(image, filename);
           iw.addDocument(document);
           count++;
           if (count % 50 == 0)
               System.out.println("finished " + (count * 100) / allImages.size() + "% of the images.");
       }
       iw.close();
       return document;
   }
 

import net.semanticmetadata.lire.utils.LuceneUtils; //导入依赖的package包/类
public void testIndexing() throws IOException {
        List<String> allImages = FileUtils.getAllImages(new File(filePath), true);
        System.out.println("Found " + allImages.size() + " files.");
        IndexWriter iw = LuceneUtils.createIndexWriter(indexPath, true);
//        ParallelIndexer pix = new ParallelIndexer(allImages, new MirFlickrDocumentBuilder());
        ParallelIndexer pix = new ParallelIndexer(allImages, new MetadataBuilder());
        new Thread(pix).start();
        Document doc;
        javax.swing.ProgressMonitor pm = new javax.swing.ProgressMonitor(null, "Indexing ...", "", 0, allImages.size());
        int count = 0;
        while ((doc = pix.getNext()) != null) {
            iw.addDocument(doc);
            pm.setProgress(++count);
            pm.setNote(count + " documents finished");
        }
        iw.close();
    }