groovy pdfbox 识别二维码

使用PDFBox识别二维码需要以下步骤：

1.导入相关库文件

import org.apache.pdfbox.pdmodel.PDDocument;
import org.apache.pdfbox.pdmodel.PDPage;
import org.apache.pdfbox.pdmodel.common.PDStream;
import org.apache.pdfbox.text.PDFTextStripper;
import org.apache.pdfbox.tools.imageio.ImageIOUtil;
import org.apache.pdfbox.contentstream.operator.Operator;
import org.apache.pdfbox.cos.COSBase;
import org.apache.pdfbox.cos.COSName;
import org.apache.pdfbox.cos.COSStream;
import org.apache.pdfbox.filter.Filter;
import org.apache.pdfbox.filter.FilterFactory;
import org.apache.pdfbox.filter.FilterManager;
import org.apache.pdfbox.filter.FilterOperator;
import org.apache.pdfbox.filter.MultiplyFilter;
import org.apache.pdfbox.filter.MultiplyNonStrokingCA;
import org.apache.pdfbox.filter.MultiplyNonStrokingCS;
import org.apache.pdfbox.filter.MultiplyNonStrokingCMYK;
import org.apache.pdfbox.filter.MultiplyNonStrokingGray;
import org.apache.pdfbox.filter.MultiplyNonStrokingRGB;
import org.apache.pdfbox.filter.MultiplyStrokingCA;
import org.apache.pdfbox.filter.MultiplyStrokingCS;
import org.apache.pdfbox.filter.MultiplyStrokingCMYK;
import org.apache.pdfbox.filter.MultiplyStrokingGray;
import org.apache.pdfbox.filter.MultiplyStrokingRGB;
import org.apache.pdfbox.filter.MultiplyAll;
import org.apache.pdfbox.filter.MultiplyAllNonStroking;
import org.apache.pdfbox.filter.MultiplyAllStroking;
import org.apache.pdfbox.filter.MultiplyAllNonStrokingCMYK;
import org.apache.pdfbox.filter.MultiplyAllStrokingCMYK;
import org.apache.pdfbox.filter.MultiplyAllNonStrokingGray;
import org.apache.pdfbox.filter.MultiplyAllStrokingGray;
import org.apache.pdfbox.filter.MultiplyAllNonStrokingRGB;
import org.apache.pdfbox.filter.MultiplyAllStrokingRGB;
import org.apache.pdfbox.filter.MultiplyAllNonStrokingCA;
import org.apache.pdfbox.filter.MultiplyAllStrokingCA;
import org.apache.pdfbox.filter.MultiplyAllNonStrokingCS;
import org.apache.pdfbox.filter.MultiplyAllStrokingCS;
import org.apache.pdfbox.filter.MultiplyAllStrokingNone;
import org.apache.pdfbox.filter.MultiplyAllNonStrokingNone;
import org.apache.pdfbox.filter.MultiplyAllStrokingPattern;
import org.apache.pdfbox.filter.MultiplyAllNonStrokingPattern;

2.读取PDF文件

PDDocument document = PDDocument.load(new File("input.pdf"));

3.获取PDF页面

PDPage page = document.getPage(0);

4.获取页面内容

PDFTextStripper stripper = new PDFTextStripper();
String text = stripper.getText(document);

5.遍历页面内容，查找二维码

List<COSBase> arguments = new ArrayList<COSBase>();
for (int i = 0; i < page.getContents().getOperators().size(); i++) {
    Operator operator = page.getContents().getOperators().get(i);
    if (operator.getName().equals("q")) {
        COSStream stream = (COSStream) page.getContents().get(i - 1);
        Filter filter = FilterFactory.INSTANCE.getFilter(stream);
        PDStream pdStream = new PDStream(stream);
        pdStream.addCompression();
        pdStream.setFilters(filter);
        pdStream.setDecodeParms(stream.getDecodeParams());
        pdStream.writeTo(new ByteArrayOutputStream());
        COSName filterName = filter.getName();
        FilterManager manager = FilterFactory.INSTANCE.getFilterManager();
        FilterOperator filterOperator = manager.getFilterOperator(filterName);
        if (filterOperator instanceof MultiplyFilter) {
            arguments.add(filterName);
        }
    }
}

6.将二维码提取出来

for (COSBase argument : arguments) {
    if (argument instanceof COSName) {
        COSName cosName = (COSName) argument;
        if (cosName.getName().equals("DCTDecode")) {
            // 提取JPEG二维码
            PDStream stream = page.getContents().getImages().get(0).getStream();
            ImageIOUtil.writeImage(stream.getByteArray(), "output.jpg", 300);
        } else if (cosName.getName().equals("JPXDecode")) {
            // 提取JPEG2000二维码
            PDStream stream = page.getContents().getImages().get(0).getStream();
            ImageIOUtil.writeImage(stream.getByteArray(), "output.jp2", 300);
        } else if (cosName.getName().equals("CCITTFaxDecode")) {
            // 提取CCITT二维码
            PDStream stream = page.getContents().getImages().get(0).getStream();
            ImageIOUtil.writeImage(stream.getByteArray(), "output.tif", 300);
        } else if (cosName.getName().equals("JBIG2Decode")) {
            // 提取JBIG2二维码
            PDStream stream = page.getContents().getImages().get(0).getStream();
            ImageIOUtil.writeImage(stream.getByteArray(), "output.jb2", 300);
        } else if (cosName.getName().equals("RunLengthDecode")) {
            // 提取RLE二维码
            PDStream stream = page.getContents().getImages().get(0).getStream();
            ImageIOUtil.writeImage(stream.getByteArray(), "output.rle", 300);
        } else if (cosName.getName().equals("FlateDecode")) {
            // 提取ZLIB二维码
            PDStream stream = page.getContents().getImages().get(0).getStream();
            ImageIOUtil.writeImage(stream.getByteArray(), "output.png", 300);
        } else if (cosName.getName().equals("LZWDecode")) {
            // 提取LZW二维码
            PDStream stream = page.getContents().getImages().get(0).getStream();
            ImageIOUtil.writeImage(stream.getByteArray(), "output.tif", 300);
        }
    }
}

7.关闭PDF文件

document.close();
``