将 pdf 页面渲染为图像 PDFBox Java 时缺少 PDAnnotationTextMarkup

如何解决将 pdf 页面渲染为图像 PDFBox Java 时缺少 PDAnnotationTextMarkup

我使用 PDFBox API 编写的代码可以突出显示 PDF 中的单词，但是当我将突出显示的 PDF 页面转换为图像时，我突出显示的任何内容都会从图像中消失。

下面的屏幕截图带有突出显示的文本，为了突出显示，我使用了 PDFBox 的 PDAnnotationTextMarkup 类：

Highlighted PDF Page

下面是pdf页面转图片后的图片：

Highlighted PDF Page Image after converting

以下是我用于将 PDF 转换为图像的代码：

PDDocument document = PDDocument.load(new File(pdfFilename));
PDFRenderer pdfRenderer = new PDFRenderer(document);
int pageCounter = 0;
for (PDPage page : document.getPages())
{
   BufferedImage bim = pdfRenderer.renderImageWithDPI(pageCounter,300,ImageType.RGB);

   ImageIOUtil.writeImage(bim,pdfFilename + "-" + (pageCounter++) + ".png",300);
}
document.close();

请指出这里有什么问题，为什么 PDFRenderer 无法获取带有突出显示的红色框的 PDF 页面图像。

以下是我用来突出显示 PDF 文本的代码：

private void highlightText(String pdfFilePath,String highlightedPdfFilePath) {
    try {
        // Loading an existing document
        
        File file = new File(highlightedPdfFilePath);
        if (!file.exists()) {
            file = new File(pdfFilePath);
        }
        
        PDDocument document = PDDocument.load(file);
        
        // extended PDFTextStripper class
        PDFTextStripper stripper = new PDFTextHighlighter();
        
        // Get number of pages
        int number_of_pages = document.getDocumentCatalog().getPages().getCount();

        // The method writeText will invoke an override version of
        // writeString
        Writer dummy = new OutputStreamWriter(new ByteArrayOutputStream());
        stripper.writeText(document,dummy);
        
        // Print collected information
        System.out.println("tokenStream:::"+tokenStream);
        System.out.println("tokenStream size::"+tokenStream.size());
        System.out.println("coordinates size::"+coordinates.size());

        double page_height;
        double page_width;
        double width,height,minx,maxx,miny,maxy;
        int rotation;
        
        // scan each page and highlitht all the words inside them
        for (int page_index = 0; page_index < number_of_pages; page_index++) {
            // get current page
            PDPage page = document.getPage(page_index);
            
            // Get annotations for the selected page
            List<PDAnnotation> annotations = page.getAnnotations();

            // Define a color to use for highlighting text
            PDColor red = new PDColor(new float[] { 1,0 },PDDeviceRGB.INSTANCE);

            // Page height and width
            page_height = page.getMediaBox().getHeight();
            page_width = page.getMediaBox().getWidth();
            // Scan collected coordinates
            for (int i = 0; i < coordinates.size(); i++) {
                if (!differencePgaeNumber.contains(page_index)) {
                    differencePgaeNumber.add(page_index);
                }
                // if the current coordinates are not related to the current
                // page,ignore them
                if ((int) coordinates.get(i)[4] != (page_index + 1))
                    continue;
                else {
                    // get rotation of the page...portrait..landscape..
                    rotation = (int) coordinates.get(i)[7];

                    // page rotated of 90degrees
                    if (rotation == 90) {
                        height = coordinates.get(i)[5];
                        width = coordinates.get(i)[6];
                        width = (page_height * width) / page_width;

                        // define coordinates of a rectangle
                        maxx = coordinates.get(i)[1];
                        minx = coordinates.get(i)[1] - height;
                        miny = coordinates.get(i)[0];
                        maxy = coordinates.get(i)[0] + width;
                    } else // i should add here the cases -90/-180 degrees
                    {
                        height = coordinates.get(i)[5];
                        minx = coordinates.get(i)[0];
                        maxx = coordinates.get(i)[2];
                        miny = page_height - coordinates.get(i)[1];
                        maxy = page_height - coordinates.get(i)[3] + height;
                    }
                    
                    // Add an annotation for each scanned word
                    PDAnnotationTextMarkup txtMark = new PDAnnotationTextMarkup(
                            PDAnnotationTextMarkup.SUB_TYPE_HIGHLIGHT);
                    txtMark.setColor(red);
                    txtMark.setConstantOpacity((float) 0.3); // 30%
                                                                // transparent
                    PDRectangle position = new PDRectangle();
                    position.setLowerLeftX((float) minx);
                    position.setLowerLeftY((float) miny);
                    position.setUpperRightX((float) maxx);
                    position.setUpperRightY((float) ((float) maxy + height));
                    txtMark.setRectangle(position);

                    float[] quads = new float[8];
                    quads[0] = position.getLowerLeftX(); // x1
                    quads[1] = position.getUpperRightY() - 2; // y1
                    quads[2] = position.getUpperRightX(); // x2
                    quads[3] = quads[1]; // y2
                    quads[4] = quads[0]; // x3
                    quads[5] = position.getLowerLeftY() - 2; // y3
                    quads[6] = quads[2]; // x4
                    quads[7] = quads[5]; // y5
                    txtMark.setQuadPoints(quads);
                    txtMark.setContents(tokenStream.get(i).toString());
                    
                    annotations.add(txtMark);
                }
            }
        }

        // Saving the document in a new file
        File highlighted_doc = new File(highlightedPdfFilePath);
        
        document.save(highlighted_doc);
        document.close();
    } catch (IOException e) {
        System.out.println(e);
    }
}

解决方法

您需要使用此调用构建注释的视觉外观：

txtMark.constructAppearances(document);