Java Histogram类代码示例

import net.semanticmetadata.lire.imageanalysis.Histogram; //导入依赖的package包/类
/**
 * Takes one single document and creates the visual words and adds them to the document. The same document is returned.
 *
 * @param d the document to use for adding the visual words
 * @return
 * @throws IOException
 */
public Document getVisualWords(Document d) throws IOException {
    clusters = Cluster.readClusters(clusterFile);
    int[] tmpHist = new int[clusters.length];
    LireFeature f = getFeatureInstance();
    IndexableField[] fields = d.getFields(localFeatureFieldName);
    // find the appropriate cluster for each feature:
    for (int j = 0; j < fields.length; j++) {
        f.setByteArrayRepresentation(fields[j].binaryValue().bytes, fields[j].binaryValue().offset, fields[j].binaryValue().length);
        tmpHist[clusterForFeature((Histogram) f)]++;
    }
    normalize(tmpHist);
    d.add(new TextField(visualWordsFieldName, arrayToVisualWordString(tmpHist), Field.Store.YES));
    d.add(new StringField(localFeatureHistFieldName, SerializationUtils.arrayToString(tmpHist), Field.Store.YES));
    d.removeFields(localFeatureFieldName);
    return d;
}
 

import net.semanticmetadata.lire.imageanalysis.Histogram; //导入依赖的package包/类
/**
     * Takes one single document and creates the visual words and adds them to the document. The same document is returned.
     *
     * @param d the document to use for adding the visual words
     * @return
     * @throws java.io.IOException
     */
    public Document getVisualWords(Document d) throws IOException {  // TODO: Adapt to VLAD!
        clusters = Cluster.readClusters(clusterFile);
        double[] vlad = null;
        LireFeature f = getFeatureInstance();
        IndexableField[] fields = d.getFields(localFeatureFieldName);
        // find the appropriate cluster for each feature:
        for (int j = 0; j < fields.length; j++) {
            f.setByteArrayRepresentation(fields[j].binaryValue().bytes, fields[j].binaryValue().offset, fields[j].binaryValue().length);
            if (vlad == null) {
                vlad = new double[clusters.length * f.getDoubleHistogram().length];
                Arrays.fill(vlad, 0d);
            }
            int clusterIndex = clusterForFeature((Histogram) f);
            double[] mean = clusters[clusterIndex].getMean();
            for (int k = 0; k < f.getDoubleHistogram().length; k++) {
                vlad[clusterIndex * f.getDoubleHistogram().length + k] += f.getDoubleHistogram()[k] - mean[k];
            }

        }
        normalize(vlad);
        GenericByteLireFeature feat = new GenericByteLireFeature();
        feat.setData(vlad);
        d.add(new StoredField(vladFieldName, feat.getByteArrayRepresentation()));
//        d.add(new StringField(localFeatureHistFieldName, SerializationUtils.arrayToString(tmpHist), Field.Store.YES));
        d.removeFields(localFeatureFieldName);
        return d;
    }
 

import net.semanticmetadata.lire.imageanalysis.Histogram; //导入依赖的package包/类
/**
 * Find the appropriate cluster for a given feature.
 *
 * @param f
 * @return the index of the cluster.
 */
private int clusterForFeature(Histogram f) {
    double distance = clusters[0].getDistance(f);
    double tmp;
    int result = 0;
    for (int i = 1; i < clusters.length; i++) {
        tmp = clusters[i].getDistance(f);
        if (tmp < distance) {
            distance = tmp;
            result = i;
        }
    }
    return result;
}
 

import net.semanticmetadata.lire.imageanalysis.Histogram; //导入依赖的package包/类
public void indexMissing() throws IOException {
        // Reading clusters from disk:
        clusters = Cluster.readClusters(clusterFile);
        //  create & store histograms:
        System.out.println("Creating histograms ...");
        int[] tmpHist = new int[numClusters];
        LireFeature f = getFeatureInstance();
        // based on bug report from Einav Itamar <[email protected]>
        IndexWriter iw = LuceneUtils.createIndexWriter(((DirectoryReader) reader).directory(),
                false, LuceneUtils.AnalyzerType.WhitespaceAnalyzer);
        for (int i = 0; i < reader.maxDoc(); i++) {
//            if (!reader.isDeleted(i)) {
            for (int j = 0; j < tmpHist.length; j++) {
                tmpHist[j] = 0;
            }
            Document d = reader.document(i);
            // Only if there are no values yet:
            if (d.getValues(visualWordsFieldName) == null || d.getValues(visualWordsFieldName).length == 0) {
                IndexableField[] fields = d.getFields(localFeatureFieldName);
                // find the appropriate cluster for each feature:
                for (int j = 0; j < fields.length; j++) {
                    f.setByteArrayRepresentation(fields[j].binaryValue().bytes, fields[j].binaryValue().offset, fields[j].binaryValue().length);
                    tmpHist[clusterForFeature((Histogram) f)]++;
                }
                normalize(tmpHist);
                d.add(new TextField(visualWordsFieldName, arrayToVisualWordString(tmpHist), Field.Store.YES));
                d.add(new StringField(localFeatureHistFieldName, SerializationUtils.arrayToString(tmpHist), Field.Store.YES));
                // now write the new one. we use the identifier to update ;)
                iw.updateDocument(new Term(DocumentBuilder.FIELD_NAME_IDENTIFIER, d.getValues(DocumentBuilder.FIELD_NAME_IDENTIFIER)[0]), d);
            }
//            }
        }
        iw.commit();
        iw.close();
        System.out.println("Finished.");
    }
 

import net.semanticmetadata.lire.imageanalysis.Histogram; //导入依赖的package包/类
public double getDistance(Histogram f) {
    return getDistance(f.descriptor);
}
 

import net.semanticmetadata.lire.imageanalysis.Histogram; //导入依赖的package包/类
/**
     * Indexes all documents in the index, that do not include the VLAD feature yet.
     *
     * @throws IOException
     */
    public void indexMissing() throws IOException {
        // Reading clusters from disk:
        clusters = Cluster.readClusters(clusterFile);
        //  create & store histograms:
        System.out.println("Creating histograms ...");
//        int[] tmpHist = new int[numClusters];
        LireFeature f = getFeatureInstance();
        IndexWriter iw = LuceneUtils.createIndexWriter(((DirectoryReader) reader).directory(), true, LuceneUtils.AnalyzerType.WhitespaceAnalyzer);
        for (int i = 0; i < reader.maxDoc(); i++) {
//            if (!reader.isDeleted(i)) {
            Document d = reader.document(i);
            double[] vlad = null;

            // Only if there are no values yet:
            if (d.getValues(vladFieldName) == null || d.getValues(vladFieldName).length == 0) {
                IndexableField[] fields = d.getFields(localFeatureFieldName);
                // find the appropriate cluster for each feature:
                for (int j = 0; j < fields.length; j++) {
                    f.setByteArrayRepresentation(fields[j].binaryValue().bytes, fields[j].binaryValue().offset, fields[j].binaryValue().length);
                    if (vlad == null) {  // init vlad if it is null.
                        vlad = new double[clusters.length * f.getDoubleHistogram().length];
                        Arrays.fill(vlad, 0d);
                    }
                    int clusterIndex = clusterForFeature((Histogram) f);
                    double[] mean = clusters[clusterIndex].getMean();
                    for (int k = 0; k < f.getDoubleHistogram().length; k++) {
                        vlad[clusterIndex * f.getDoubleHistogram().length + k] += f.getDoubleHistogram()[k] - mean[k];
                    }

                }
                normalize(vlad);
                GenericByteLireFeature feat = new GenericByteLireFeature();
                feat.setData(vlad);
//                System.out.println(feat.getStringRepresentation());
                d.add(new StoredField(vladFieldName, feat.getByteArrayRepresentation()));
                iw.updateDocument(new Term(DocumentBuilder.FIELD_NAME_IDENTIFIER, d.getValues(DocumentBuilder.FIELD_NAME_IDENTIFIER)[0]), d);
            }
//            }
        }
        iw.commit();
        iw.close();
        System.out.println("Finished.");
    }
 

import net.semanticmetadata.lire.imageanalysis.Histogram; //导入依赖的package包/类
public void run() {
//            int[] tmpHist = new int[numClusters];
            LireFeature f = getFeatureInstance();
            for (int i = start; i < end; i++) {
                try {
//                    if (!reader.isDeleted(i)) {    // TODO!
//                    for (int j = 0; j < tmpHist.length; j++) {
//                        tmpHist[j] = 0;
//                    }
                    Document d = reader.document(i);
                    IndexableField[] fields = d.getFields(localFeatureFieldName);
                    // remove the fields if they are already there ...
                    d.removeField(vladFieldName);
//                    d.removeField(localFeatureHistFieldName);
                    double[] vlad = null;

                    // VLAD - Vector of Locally Aggregated Descriptors
                    for (int j = 0; j < fields.length; j++) {
                        f.setByteArrayRepresentation(fields[j].binaryValue().bytes, fields[j].binaryValue().offset, fields[j].binaryValue().length);
                        if (vlad == null) {  // init vlad if it is null.
                            vlad = new double[clusters.length * f.getDoubleHistogram().length];
                            Arrays.fill(vlad, 0d);
                        }
                        int clusterIndex = clusterForFeature((Histogram) f);
//                        System.out.println("clusterIndex = " + clusterIndex);
                        double[] mean = clusters[clusterIndex].getMean();
                        for (int k = 0; k < f.getDoubleHistogram().length; k++) {
//                            System.out.println((clusterIndex*f.getDoubleHistogram().length+k) + " - mean: " + mean.length + " - feature: " + f.getDoubleHistogram().length);
                            vlad[clusterIndex * f.getDoubleHistogram().length + k] += f.getDoubleHistogram()[k] - mean[k];
                        }
                    }
                    normalize(vlad);
                    GenericByteLireFeature feat = new GenericByteLireFeature();
                    feat.setData(vlad);
//                    System.out.println(feat.getStringRepresentation());
                    d.add(new StoredField(vladFieldName, feat.getByteArrayRepresentation()));
//                    d.add(new StringField(localFeatureHistFieldName, SerializationUtils.arrayToString(tmpHist), Field.Store.YES));

                    // remove local features to save some space if requested:
                    if (DELETE_LOCAL_FEATURES) {
                        d.removeFields(localFeatureFieldName);
                    }
                    // now write the new one. we use the identifier to update ;)
                    iw.updateDocument(new Term(DocumentBuilder.FIELD_NAME_IDENTIFIER, d.getValues(DocumentBuilder.FIELD_NAME_IDENTIFIER)[0]), d);
                    if (pm != null) {
                        double len = (double) (end - start);
                        double percent = (double) (i - start) / len * 45d + 50;
                        pm.setProgress((int) percent);
                        pm.setNote("Creating visual words, ~" + (int) percent + "% finished");
                    }
//                    }
                } catch (IOException e) {
                    e.printStackTrace();
                }
            }
        }
 

import net.semanticmetadata.lire.imageanalysis.Histogram; //导入依赖的package包/类
public void run() {
            int[] tmpHist = new int[numClusters];
            LireFeature f = getFeatureInstance();
            for (int i = start; i < end; i++) {
                try {
//                    if (!reader.isDeleted(i)) {    // TODO!
                    for (int j = 0; j < tmpHist.length; j++) {
                        tmpHist[j] = 0;
                    }
                    Document d = reader.document(i);
                    IndexableField[] fields = d.getFields(localFeatureFieldName);
                    // remove the fields if they are already there ...
                    d.removeField(visualWordsFieldName);
                    d.removeField(localFeatureHistFieldName);

                    // find the appropriate cluster for each feature:
                    for (int j = 0; j < fields.length; j++) {
                        f.setByteArrayRepresentation(fields[j].binaryValue().bytes, fields[j].binaryValue().offset, fields[j].binaryValue().length);
                        tmpHist[clusterForFeature((Histogram) f)]++;
                    }
                    normalize(tmpHist);
                    d.add(new TextField(visualWordsFieldName, arrayToVisualWordString(tmpHist), Field.Store.YES));
                    d.add(new StringField(localFeatureHistFieldName, SerializationUtils.arrayToString(tmpHist), Field.Store.YES));

                    // remove local features to save some space if requested:
                    if (DELETE_LOCAL_FEATURES) {
                        d.removeFields(localFeatureFieldName);
                    }
                    // now write the new one. we use the identifier to update ;)
                    iw.updateDocument(new Term(DocumentBuilder.FIELD_NAME_IDENTIFIER, d.getValues(DocumentBuilder.FIELD_NAME_IDENTIFIER)[0]), d);
                    if (pm != null) {
                        double len = (double) (end - start);
                        double percent = (double) (i - start) / len * 45d + 50;
                        pm.setProgress((int) percent);
                        pm.setNote("Creating visual words, ~" + (int) percent + "% finished");
                    }
//                    }
                } catch (IOException e) {
                    e.printStackTrace();
                }
            }
        }
 

import net.semanticmetadata.lire.imageanalysis.Histogram; //导入依赖的package包/类
/**
 * Used to find the cluster of a feature actually used in the clustering process (so
 * it is known by the k-means class).
 *
 * @param f the feature to search for
 * @return the index of the Cluster
 */
public int getClusterOfFeature(Histogram f) {
    if (featureIndex == null) createIndex();
    return featureIndex.get(f);
}