这应该做的伎俩:
public static void main(String[] args) throws IOException {
PDDocument document = PDDocument.load(new File(""));
int numImages = 0;
for (int i = 0; i < document.getNumberOfPages(); i++)
{
PDPage page = document.getPage(i);
CountImages countImages = new CountImages(page);
countImages.processPage(page);
numImages += countImages.numImages;
}
System.out.println(numImages);
}
static class CountImages extends PDFGraphicsStreamEngine {
public int numImages = 0;
private final Set<COSStream> duplicates = new HashSet<>();
protected CountImages(PDPage page) throws IOException
{
super(page);
}
@Override
public void appendRectangle(Point2D pd, Point2D pd1, Point2D pd2, Point2D pd3) throws IOException {
}
@Override
public void drawImage(PDImage pdImage) throws IOException {
if (pdImage instanceof PDImageXObject) {
PDImageXObject xobject = (PDImageXObject)pdImage;
if (duplicates.contains(xobject.getCOSObject()) == false) {
numImages++;
duplicates.add(xobject.getCOSObject());
}
} else {
numImages++; //means its an inline image
}
}
@Override
public void clip(int i) throws IOException {
}
@Override
public void moveTo(float f, float f1) throws IOException {
}
@Override
public void lineTo(float f, float f1) throws IOException {
}
@Override
public void curveTo(float f, float f1, float f2, float f3, float f4, float f5) throws IOException {
}
@Override
public Point2D getCurrentPoint() throws IOException {
return new Point2D.Float(0, 0);
}
@Override
public void closePath() throws IOException {
}
@Override
public void endPath() throws IOException {
}
@Override
public void strokePath() throws IOException {
}
@Override
public void fillPath(int i) throws IOException {
}
@Override
public void fillAndStrokePath(int i) throws IOException {
}
@Override
public void shadingFill(COSName cosn) throws IOException {
}
}
这忽略相当多的地方图像即可。此外,不保证页面资源中的图像实际上用于页面。 – mkl
@mkl有趣但含糊。你为什么不分享你的智慧珍珠,并发表更好的答案?我只用两种测试用例来确保特定的PDF包含图像或不包含图像。由于它的工作可靠,我没有深入探讨这个话题。 –
@Würgspaß在源代码下载或https://svn.apache.org/viewvc/pdfbox/trunk/tools/src/main/java/org/apache/pdfbox/tools/中查看ExtractImages源代码ExtractImages.java?view=markup –