Extract images using iTextSharp

Question

I have been using this code with great success to pull out the first image found in each page of a PDF. However, it is now not working with some new PDFs for an uknown reason. I have used other tools (Datalogics, etc) that do pull out the images fine with these new PDFs. However, I do not want to buy Datalogics or any tool if I can use iTextSharp. Can anybody tell me why this code is not finding the images in the PDF?

Knowns: my PDFs only have 1 image per page and nothing else.

using iTextSharp.text;
using iTextSharp.text.pdf;
...
public static void ExtractImagesFromPDF(string sourcePdf, string outputPath)
{
    // NOTE:  This will only get the first image it finds per page.
    PdfReader pdf = new PdfReader(sourcePdf);
    RandomAccessFileOrArray raf = new iTextSharp.text.pdf.RandomAccessFileOrArray(sourcePdf);

    try
    {
        for (int pageNumber = 1; pageNumber <= pdf.NumberOfPages; pageNumber++)
        {
            PdfDictionary pg = pdf.GetPageN(pageNumber);
            PdfDictionary res = (PdfDictionary)PdfReader.GetPdfObject(pg.Get(PdfName.RESOURCES));

            PdfDictionary xobj = (PdfDictionary)PdfReader.GetPdfObject(res.Get(PdfName.XOBJECT));
            if (xobj != null)
            {
                foreach (PdfName name in xobj.Keys)
                {
                    PdfObject obj = xobj.Get(name);
                    if (obj.IsIndirect())
                    {
                        PdfDictionary tg = (PdfDictionary)PdfReader.GetPdfObject(obj);
                        PdfName type = (PdfName)PdfReader.GetPdfObject(tg.Get(PdfName.SUBTYPE));
                        if (PdfName.IMAGE.Equals(type))
                        {
                            int XrefIndex = Convert.ToInt32(((PRIndirectReference)obj).Number.ToString(System.Globalization.CultureInfo.InvariantCulture));
                            PdfObject pdfObj = pdf.GetPdfObject(XrefIndex);
                            PdfStream pdfStrem = (PdfStream)pdfObj;
                            byte[] bytes = PdfReader.GetStreamBytesRaw((PRStream)pdfStrem);
                            if ((bytes != null))
                            {
                                using (System.IO.MemoryStream memStream = new System.IO.MemoryStream(bytes))
                                {
                                    memStream.Position = 0;
                                    System.Drawing.Image img = System.Drawing.Image.FromStream(memStream);
                                    // must save the file while stream is open.
                                    if (!Directory.Exists(outputPath))
                                        Directory.CreateDirectory(outputPath);

                                    string path = Path.Combine(outputPath, String.Format(@"{0}.jpg", pageNumber));
                                    System.Drawing.Imaging.EncoderParameters parms = new System.Drawing.Imaging.EncoderParameters(1);
                                    parms.Param[0] = new System.Drawing.Imaging.EncoderParameter(System.Drawing.Imaging.Encoder.Compression, 0);
                                    System.Drawing.Imaging.ImageCodecInfo jpegEncoder = Utilities.GetImageEncoder("JPEG");
                                    img.Save(path, jpegEncoder, parms);
                                    break;
                                }
                            }
                        }
                    }
                }
            }
        }
    }

    catch
    {
        throw;
    }
    finally
    {
        pdf.Close();
        raf.Close();
    }
}

Answer 1

I found that my problem was that I was not recursively searching inside of forms and groups for images. Basically, the original code would only find images that were embedded at the root of the pdf document. Here is the revised method plus a new method (FindImageInPDFDictionary) that recursively searches for images in the page. NOTE: the flaws of only supporting JPEG and non-compressed images still applies. See R Ubben's code for options to fix those flaws. HTH someone.

    public static void ExtractImagesFromPDF(string sourcePdf, string outputPath)
    {
        // NOTE:  This will only get the first image it finds per page.
        PdfReader pdf = new PdfReader(sourcePdf);
        RandomAccessFileOrArray raf = new iTextSharp.text.pdf.RandomAccessFileOrArray(sourcePdf);

        try
        {
            for (int pageNumber = 1; pageNumber <= pdf.NumberOfPages; pageNumber++)
            {
                PdfDictionary pg = pdf.GetPageN(pageNumber);

                // recursively search pages, forms and groups for images.
                PdfObject obj = FindImageInPDFDictionary(pg);
                if (obj != null)
                {

                    int XrefIndex = Convert.ToInt32(((PRIndirectReference)obj).Number.ToString(System.Globalization.CultureInfo.InvariantCulture));
                    PdfObject pdfObj = pdf.GetPdfObject(XrefIndex);
                    PdfStream pdfStrem = (PdfStream)pdfObj;
                    byte[] bytes = PdfReader.GetStreamBytesRaw((PRStream)pdfStrem);
                    if ((bytes != null))
                    {
                        using (System.IO.MemoryStream memStream = new System.IO.MemoryStream(bytes))
                        {
                            memStream.Position = 0;
                            System.Drawing.Image img = System.Drawing.Image.FromStream(memStream);
                            // must save the file while stream is open.
                            if (!Directory.Exists(outputPath))
                                Directory.CreateDirectory(outputPath);

                            string path = Path.Combine(outputPath, String.Format(@"{0}.jpg", pageNumber));
                            System.Drawing.Imaging.EncoderParameters parms = new System.Drawing.Imaging.EncoderParameters(1);
                            parms.Param[0] = new System.Drawing.Imaging.EncoderParameter(System.Drawing.Imaging.Encoder.Compression, 0);
                            System.Drawing.Imaging.ImageCodecInfo jpegEncoder = Utilities.GetImageEncoder("JPEG");
                            img.Save(path, jpegEncoder, parms);
                        }
                    }
                }
            }
        }
        catch
        {
            throw;
        }
        finally
        {
            pdf.Close();
            raf.Close();
        }


    }

     private static PdfObject FindImageInPDFDictionary(PdfDictionary pg)
    {
        PdfDictionary res =
            (PdfDictionary)PdfReader.GetPdfObject(pg.Get(PdfName.RESOURCES));


        PdfDictionary xobj =
          (PdfDictionary)PdfReader.GetPdfObject(res.Get(PdfName.XOBJECT));
        if (xobj != null)
        {
            foreach (PdfName name in xobj.Keys)
            {

                PdfObject obj = xobj.Get(name);
                if (obj.IsIndirect())
                {
                    PdfDictionary tg = (PdfDictionary)PdfReader.GetPdfObject(obj);

                    PdfName type =
                      (PdfName)PdfReader.GetPdfObject(tg.Get(PdfName.SUBTYPE));

                    //image at the root of the pdf
                    if (PdfName.IMAGE.Equals(type))
                    {
                        return obj;
                    }// image inside a form
                    else if (PdfName.FORM.Equals(type))
                    {
                        return FindImageInPDFDictionary(tg);
                    } //image inside a group
                    else if (PdfName.GROUP.Equals(type))
                    {
                        return FindImageInPDFDictionary(tg);
                    }

                }
            }
        }

        return null;

    }

Answer 2

The following code incorporates all of Dave and R Ubben's ideas above, plus it returns a full list of all the images and also deals with multiple bit depths. I had to convert it to VB for the project I'm working on though, sorry about that...

Private Sub getAllImages(ByVal dict As pdf.PdfDictionary, ByVal images As List(Of Byte()), ByVal doc As pdf.PdfReader)
    Dim res As pdf.PdfDictionary = CType(pdf.PdfReader.GetPdfObject(dict.Get(pdf.PdfName.RESOURCES)), pdf.PdfDictionary)
    Dim xobj As pdf.PdfDictionary = CType(pdf.PdfReader.GetPdfObject(res.Get(pdf.PdfName.XOBJECT)), pdf.PdfDictionary)

    If xobj IsNot Nothing Then
        For Each name As pdf.PdfName In xobj.Keys
            Dim obj As pdf.PdfObject = xobj.Get(name)
            If (obj.IsIndirect) Then
                Dim tg As pdf.PdfDictionary = CType(pdf.PdfReader.GetPdfObject(obj), pdf.PdfDictionary)
                Dim subtype As pdf.PdfName = CType(pdf.PdfReader.GetPdfObject(tg.Get(pdf.PdfName.SUBTYPE)), pdf.PdfName)
                If pdf.PdfName.IMAGE.Equals(subtype) Then
                    Dim xrefIdx As Integer = CType(obj, pdf.PRIndirectReference).Number
                    Dim pdfObj As pdf.PdfObject = doc.GetPdfObject(xrefIdx)
                    Dim str As pdf.PdfStream = CType(pdfObj, pdf.PdfStream)
                    Dim bytes As Byte() = pdf.PdfReader.GetStreamBytesRaw(CType(str, pdf.PRStream))

                    Dim filter As String = tg.Get(pdf.PdfName.FILTER).ToString
                    Dim width As String = tg.Get(pdf.PdfName.WIDTH).ToString
                    Dim height As String = tg.Get(pdf.PdfName.HEIGHT).ToString
                    Dim bpp As String = tg.Get(pdf.PdfName.BITSPERCOMPONENT).ToString

                    If filter = "/FlateDecode" Then
                        bytes = pdf.PdfReader.FlateDecode(bytes, True)
                        Dim pixelFormat As System.Drawing.Imaging.PixelFormat
                        Select Case Integer.Parse(bpp)
                            Case 1
                                pixelFormat = Drawing.Imaging.PixelFormat.Format1bppIndexed
                            Case 24
                                pixelFormat = Drawing.Imaging.PixelFormat.Format24bppRgb
                            Case Else
                                Throw New Exception("Unknown pixel format " + bpp)
                        End Select
                        Dim bmp As New System.Drawing.Bitmap(Int32.Parse(width), Int32.Parse(height), pixelFormat)
                        Dim bmd As System.Drawing.Imaging.BitmapData = bmp.LockBits(New System.Drawing.Rectangle(0, 0, Int32.Parse(width), Int32.Parse(height)), System.Drawing.Imaging.ImageLockMode.WriteOnly, pixelFormat)
                        Marshal.Copy(bytes, 0, bmd.Scan0, bytes.Length)
                        bmp.UnlockBits(bmd)
                        Using ms As New MemoryStream
                            bmp.Save(ms, System.Drawing.Imaging.ImageFormat.Png)
                            bytes = ms.GetBuffer
                        End Using
                    End If
                    images.Add(bytes)
                ElseIf pdf.PdfName.FORM.Equals(subtype) Or pdf.PdfName.GROUP.Equals(subtype) Then
                    getAllImages(tg, images, doc)
                End If
            End If
        Next
    End If
End Sub

Answer 3

De c# version:

private IList<System.Drawing.Image> GetImagesFromPdfDict(PdfDictionary dict, PdfReader doc){
        List<System.Drawing.Image> images = new List<System.Drawing.Image>();
        PdfDictionary res = (PdfDictionary)(PdfReader.GetPdfObject(dict.Get(PdfName.RESOURCES)));
        PdfDictionary xobj = (PdfDictionary)(PdfReader.GetPdfObject(res.Get(PdfName.XOBJECT)));

        if (xobj != null)
        {
            foreach (PdfName name in xobj.Keys)
            {
                PdfObject obj = xobj.Get(name);
                if (obj.IsIndirect())
                {
                    PdfDictionary tg = (PdfDictionary)(PdfReader.GetPdfObject(obj));
                    pdf.PdfName subtype = (pdf.PdfName)(pdf.PdfReader.GetPdfObject(tg.Get(pdf.PdfName.SUBTYPE)));
                    if (pdf.PdfName.IMAGE.Equals(subtype))
                    {
                        int xrefIdx = ((pdf.PRIndirectReference)obj).Number;
                        pdf.PdfObject pdfObj = doc.GetPdfObject(xrefIdx);
                        pdf.PdfStream str = (pdf.PdfStream)(pdfObj);
                        byte[] bytes = pdf.PdfReader.GetStreamBytesRaw((pdf.PRStream)str);

                        string filter = tg.Get(pdf.PdfName.FILTER).ToString();
                        string width = tg.Get(pdf.PdfName.WIDTH).ToString();
                        string height = tg.Get(pdf.PdfName.HEIGHT).ToString();
                        string bpp = tg.Get(pdf.PdfName.BITSPERCOMPONENT).ToString();

                        if (filter == "/FlateDecode")
                        {
                            bytes = pdf.PdfReader.FlateDecode(bytes, true);
                            System.Drawing.Imaging.PixelFormat pixelFormat;
                            switch (int.Parse(bpp))
                            {
                                case 1:
                                    pixelFormat = System.Drawing.Imaging.PixelFormat.Format1bppIndexed;
                                    break;
                                case 24:
                                    pixelFormat = System.Drawing.Imaging.PixelFormat.Format24bppRgb;
                                    break;
                                default:
                                    throw new Exception("Unknown pixel format " + bpp);
                            }
                            var bmp = new System.Drawing.Bitmap(Int32.Parse(width), Int32.Parse(height), pixelFormat);
                            System.Drawing.Imaging.BitmapData bmd = bmp.LockBits(new System.Drawing.Rectangle(0, 0, Int32.Parse(width),
                                Int32.Parse(height)), System.Drawing.Imaging.ImageLockMode.WriteOnly, pixelFormat);
                            Marshal.Copy(bytes, 0, bmd.Scan0, bytes.Length);
                            bmp.UnlockBits(bmd);
                            using (var ms = new MemoryStream())
                            {
                                bmp.Save(ms, System.Drawing.Imaging.ImageFormat.Png);
                                bytes = ms.GetBuffer();
                            }
                        }
                        images.Add(System.Drawing.Image.FromStream(new MemoryStream(bytes)));
                    }
                    else if (pdf.PdfName.FORM.Equals(subtype) || pdf.PdfName.GROUP.Equals(subtype))
                    {
                        images.AddRange(GetImagesFromPdfDict(tg, doc));
                    }
                }
            }
        }
        return images;
     }

Answer 4

This is just another rehash of others' ideas, but the one that worked for me. Here I use @Malco's image grabbing snippet with R Ubben's looping:

private IList<System.Drawing.Image> GetImagesFromPdfDict(PdfDictionary dict, PdfReader doc)
{
    List<System.Drawing.Image> images = new List<System.Drawing.Image>();
    PdfDictionary res = (PdfDictionary)(PdfReader.GetPdfObject(dict.Get(PdfName.RESOURCES)));
    PdfDictionary xobj = (PdfDictionary)(PdfReader.GetPdfObject(res.Get(PdfName.XOBJECT)));

    if (xobj != null)
    {
        foreach (PdfName name in xobj.Keys)
        {
            PdfObject obj = xobj.Get(name);
            if (obj.IsIndirect())
            {
                PdfDictionary tg = (PdfDictionary)(PdfReader.GetPdfObject(obj));
                PdfName subtype = (PdfName)(PdfReader.GetPdfObject(tg.Get(PdfName.SUBTYPE)));
                if (PdfName.IMAGE.Equals(subtype))
                {
                    int xrefIdx = ((PRIndirectReference)obj).Number;
                    PdfObject pdfObj = doc.GetPdfObject(xrefIdx);
                    PdfStream str = (PdfStream)(pdfObj);

                    iTextSharp.text.pdf.parser.PdfImageObject pdfImage =
                        new iTextSharp.text.pdf.parser.PdfImageObject((PRStream)str);
                    System.Drawing.Image img = pdfImage.GetDrawingImage();

                    images.Add(img);
                }
                else if (PdfName.FORM.Equals(subtype) || PdfName.GROUP.Equals(subtype))
                {
                    images.AddRange(GetImagesFromPdfDict(tg, doc));
                }
            }
        }
    }

    return images;
}

Answer 5

The above will only work with JPEGs. Excluding inline images and embedded files, you need to go through the objects of subtype IMAGE, then look at the filter and take the appropriate action. Here's an example, assuming we have a PdfObject of subtype IMAGE:

            PdfReader pdf = new PdfReader("c:\\temp\\exp0.pdf");
        int xo=pdf.XrefSize;
        for (int i=0;i<xo;i++)
        {
            PdfObject obj=pdf.GetPdfObject(i);
            if (obj!=null && obj.IsStream())
            {
                PdfDictionary pd=(PdfDictionary)obj;
                if (pd.Contains(PdfName.SUBTYPE) && pd.Get(PdfName.SUBTYPE).ToString()=="/Image")
                {
                    string filter=pd.Get(PdfName.FILTER).ToString();
                    string width=pd.Get(PdfName.WIDTH).ToString();
                    string height=pd.Get(PdfName.HEIGHT).ToString();
                    string bpp=pd.Get(PdfName.BITSPERCOMPONENT).ToString();
                    string extent=".";
                    byte [] img=null;
                    switch (filter)
                    {
                        case "/FlateDecode":
                            byte[] arr=PdfReader.FlateDecode(PdfReader.GetStreamBytesRaw((PRStream)obj),true);
                            Bitmap bmp=new Bitmap(Int32.Parse(width),Int32.Parse(height),PixelFormat.Format24bppRgb);
                            BitmapData bmd=bmp.LockBits(new Rectangle(0,0,Int32.Parse(width),Int32.Parse(height)),ImageLockMode.WriteOnly,
                                PixelFormat.Format24bppRgb);
                            Marshal.Copy(arr,0,bmd.Scan0,arr.Length);
                            bmp.UnlockBits(bmd);
                            bmp.Save("c:\\temp\\bmp1.png",ImageFormat.Png);
                            break;
                        default:
                            break;
                    }
                }
            }
        }

This will mess the color up because of the Microsoft BGR, of course, but I wanted to keep it short. Do something similar for "/CCITTFaxDecode", etc.

Answer 6

Here is a simpler solution:

iTextSharp.text.pdf.parser.PdfImageObject pdfImage = 
                            new iTextSharp.text.pdf.parser.PdfImageObject(imgPRStream);
                        System.Drawing.Image img = pdfImage.GetDrawingImage();