PDF documents often contain valuable images—photographs, diagrams, charts, and graphics. The Extract Images demo shows how to extract all embedded images from a PDF document using PDFium VCL, saving them in optimal formats based on their characteristics.
Overview
This demo extracts embedded images (bitmaps) from PDF pages and saves them as separate image files. It includes image preview, format detection, and batch extraction with progress tracking.
Key Features
- Extract All Images – Extract every embedded image from the PDF
- Page Range Selection – Extract from specific pages only
- Smart Format Detection – Automatically choose JPEG, PNG, or BMP based on image characteristics
- Image Preview – Preview extracted images before saving
- Detailed Information – View dimensions, format, and file size
- Batch Processing – Extract multiple images with progress tracking
PDFium DLL Requirements
Before running any PDFium VCL application, ensure the PDFium DLL files are installed:
pdfium32.dll/pdfium64.dll– Standard versions (~5-6 MB)pdfium32v8.dll/pdfium64v8.dll– With V8 JavaScript engine (~23-27 MB)
Installation: Run PDFiumVCL\DLLs\CopyDlls.bat as Administrator to automatically copy the DLLs to Windows system directories.
Basic Image Extraction
Access embedded images through the Bitmap and BitmapCount properties:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 |
procedure ExtractImagesSimple; var Pdf: TPdf; I, J: Integer; Bitmap: TBitmap; begin Pdf := TPdf.Create(nil); try Pdf.FileName := 'document.pdf'; Pdf.Active := True; // Loop through all pages for I := 1 to Pdf.PageCount do begin Pdf.PageNumber := I; // Loop through all images on this page for J := 0 to Pdf.BitmapCount - 1 do begin Bitmap := Pdf.Bitmap[J]; try // Save as BMP Bitmap.SaveToFile(Format('Page%d_Image%d.bmp', [I, J + 1])); finally Bitmap.Free; end; end; end; finally Pdf.Active := False; Pdf.Free; end; end; |
Complete Extraction with Format Detection
The demo implements intelligent format selection:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 |
procedure TFormMain.ButtonExtractClick(Sender: TObject); var I, J, StartPage, EndPage: Integer; Bitmap: TBitmap; FullFileName, DetectedFormat, ImageExtension: string; ImageInfo: TImageInfo; begin FCancelled := False; FProcessedImages := 0; FTotalImages := 0; ClearExtractedImages; EnableControls(False); try Pdf.FileName := EditPdfFile.Text; Pdf.PageNumber := 0; Pdf.Active := True; ParsePageRange(EditPageRange.Text, StartPage, EndPage); if EndPage = -1 then EndPage := Pdf.PageCount; // Calculate total images for progress for I := StartPage to EndPage do begin Pdf.PageNumber := I; FTotalImages := FTotalImages + Pdf.BitmapCount; end; ProgressBar.Max := FTotalImages; // Extract images for I := StartPage to EndPage do begin if FCancelled then Break; Pdf.PageNumber := I; for J := 0 to Pdf.BitmapCount - 1 do begin if FCancelled then Break; Bitmap := Pdf.Bitmap[J]; if Assigned(Bitmap) then begin try // Detect optimal format DetectedFormat := DetectImageFormat(Bitmap); ImageExtension := GetExtensionForFormat(DetectedFormat); FullFileName := Format('%s\Page%d_Image%d%s', [FCurrentOutputDir, I, J + 1, ImageExtension]); SaveBitmapInOptimalFormat(Bitmap, FullFileName); // Store image info for preview ImageInfo.FileName := FullFileName; ImageInfo.PageNumber := I; ImageInfo.ImageIndex := J + 1; ImageInfo.Width := Bitmap.Width; ImageInfo.Height := Bitmap.Height; ImageInfo.Format := DetectedFormat; ImageInfo.Bitmap := TBitmap.Create; ImageInfo.Bitmap.Assign(Bitmap); AddImageInfo(ImageInfo); Inc(FProcessedImages); ProgressBar.Position := FProcessedImages; finally Bitmap.Free; end; end; end; end; UpdateImageList; finally Pdf.Active := False; EnableControls(True); end; end; |
Intelligent Format Detection
Choose the optimal format based on image characteristics:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 |
function TFormMain.DetectImageFormat(ABitmap: TBitmap): string; begin // Check if image has transparency (alpha channel) if ABitmap.PixelFormat = pf32bit then begin // PNG for transparency support Result := 'PNG'; end // Check if it's likely a photographic image else if (ABitmap.Width * ABitmap.Height > 100000) and (ABitmap.PixelFormat in [pf24bit, pf32bit]) then begin // Large, complex image - use JPEG for smaller file size Result := 'JPEG'; end else begin // Small or simple image - preserve quality with BMP Result := 'BMP'; end; end; function TFormMain.GetExtensionForFormat(const AFormat: string): string; begin case UpperCase(AFormat)[1] of 'J': Result := '.jpg'; 'P': Result := '.png'; 'B': Result := '.bmp'; else Result := '.bmp'; end; end; |
Saving in Optimal Format
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 |
procedure TFormMain.SaveBitmapInOptimalFormat(ABitmap: TBitmap; const AFileName: string); var JpegImg: TJPEGImage; FileExt: string; begin FileExt := UpperCase(ExtractFileExt(AFileName)); if FileExt = '.JPG' then begin // Save as JPEG with good quality JpegImg := TJPEGImage.Create; try JpegImg.Assign(ABitmap); JpegImg.CompressionQuality := 85; // Good quality/size balance JpegImg.SaveToFile(AFileName); finally JpegImg.Free; end; end else if FileExt = '.PNG' then begin // PNG would require additional library // Fall back to BMP for compatibility ABitmap.SaveToFile(ChangeFileExt(AFileName, '.bmp')); end else begin // BMP - lossless quality ABitmap.SaveToFile(AFileName); end; end; |
Using TPdfImage for Raw Image Data
For advanced use cases, access raw image data:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 |
procedure ProcessRawImageData; var Pdf: TPdf; I: Integer; PdfImage: TPdfImage; begin Pdf := TPdf.Create(nil); try Pdf.FileName := 'document.pdf'; Pdf.Active := True; Pdf.PageNumber := 1; for I := 0 to Pdf.ImageCount - 1 do begin PdfImage := Pdf.Image[I]; // Access raw image properties ShowMessage(Format('Image %d: %d x %d, %d bytes', [I, PdfImage.Width, PdfImage.Height, Length(PdfImage.Data)])); // PdfImage.Data contains raw pixel data end; finally Pdf.Active := False; Pdf.Free; end; end; |
Displaying Image Information
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 |
procedure TFormMain.UpdatePreview(Index: Integer); var Info: TImageInfo; begin if (Index >= 0) and (Index < Length(FExtractedImages)) then begin Info := FExtractedImages[Index]; // Update preview if Assigned(Info.Bitmap) then ImagePreview.Picture.Assign(Info.Bitmap); // Update info display MemoInfo.Lines.Clear; MemoInfo.Lines.Add('File: ' + ExtractFileName(Info.FileName)); MemoInfo.Lines.Add('Page: ' + IntToStr(Info.PageNumber)); MemoInfo.Lines.Add('Dimensions: ' + IntToStr(Info.Width) + ' x ' + IntToStr(Info.Height)); MemoInfo.Lines.Add('Format: ' + Info.Format); if Info.Size > 0 then MemoInfo.Lines.Add('Size: ' + FormatFloat('#,##0', Info.Size) + ' bytes'); end; end; |
Page Range Parsing
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 |
procedure TFormMain.ParsePageRange(const ARange: string; var AStartPage, AEndPage: Integer); var RangeStr: string; DashPos: Integer; begin RangeStr := Trim(ARange); AStartPage := 1; AEndPage := -1; // -1 means extract to end if (RangeStr = '') or (UpperCase(RangeStr) = 'ALL') then Exit; DashPos := Pos('-', RangeStr); if DashPos > 0 then begin // Range format: start-end AStartPage := StrToIntDef(Trim(Copy(RangeStr, 1, DashPos - 1)), 1); AEndPage := StrToIntDef(Trim(Copy(RangeStr, DashPos + 1, Length(RangeStr))), -1); end else begin // Single page AStartPage := StrToIntDef(RangeStr, 1); AEndPage := AStartPage; end; end; |
Open Extracted Images Folder
|
1 2 3 4 5 6 7 |
procedure TFormMain.ButtonOpenFolderClick(Sender: TObject); begin if DirectoryExists(FCurrentOutputDir) then ShellExecute(Handle, 'open', PChar(FCurrentOutputDir), nil, nil, SW_SHOWNORMAL) else ShowMessage('Output directory does not exist.'); end; |
Use Cases
- Digital Asset Extraction – Extract photos and graphics from marketing materials
- Document Conversion – Prepare images for web or other formats
- Archive Processing – Extract images from scanned document archives
- Content Analysis – Extract images for machine learning or analysis
Conclusion
The Extract Images demo shows how easy it is to extract embedded images from PDF documents with PDFium VCL. The component handles the complex PDF parsing while you focus on how to use the extracted images in your application.
Combined with intelligent format detection, you can build professional image extraction tools that produce optimized output for any use case.
Explore PDFium Component at loslab.com and unlock the content in your PDF documents.