Acrobat User Community Forums

Hi gyy,

If you don’t have the original file from which a PDF was created, you can save the PDF as a Word document that you can then edit in Word.

Click Export  in the Tasks toolbar, and then choose Word Document.
Click Settings to set conversion options.

Note: When you save a PDF to Word format, the resulting file isn’t equivalent to a file created in Word; some coding information may be lost.

Review these Acrobat articles:

http://kb.adobe.com/selfservice/viewCon … ;sliceId=2

http://www.adobe.com/designcenter/acrob … tepdf.html

http://www.adobe.com/designcenter/acrob … pects.html

++++++++++++++++++++++++++++++

To convert a pdf image document to a searchable image or to text and graphics, do the following:

1. With the scanned document open, select Document>Recognize Text Using OCR>Start.

Acrobat will present you with the Recognize Text dialog box.

2. Choose the radio button corresponding to the pages you want converted to a searchable image.
Most of the time, you will probably choose All pages.

3. Click the Edit button.
Acrobat will present you with the Recognize Text Settings dialog box, on the next page.

4. In the PDF Output Style popup menu, select the type of conversion you want Acrobat to perform. Your options are:

• Searchable Image (Exact) - Acrobat creates a searchable image using a very accurate algorithm to convert the text.

• Searchable Image (Compact) - Acrobat creates a searchable image, but uses a faster and slightly less-accurate ocr method.

• Formatted Text & Graphics - Acrobat replaces each full-page image with real text and line art, doing its best to match the original font.

5. Select a language to be used in the ocr conversion.
Acrobat 7 provides a list of a dozen or so languages from which you may choose.

6. Pick a resolution to which Acrobat can reduce the image when doing the ocr.
When Acrobat performs ocr, it reduces the resolution of the image to speed up the processing. The lower the downsampled resolution, the faster the processing, but the less-accurate will be the character recognition. I’d leave this at 600 dpi, myself.

6. Click the OK button to return to the Recognize Text dialog box.

7. Click the OK button to start the ocr process.

Acrobat will process the bitmapped pages.
If you chose to make a searchable image, the resulting pdf pages will look the same as they did before, but now all of the text tools will work on the apparently bitmapped text.

You can search the text, select it with the Select tool, and copy the selection. Of course, what the text tools are really operating on is the underlying text layer.
If you chose to convert the document to text and graphics, you will be looking at a document with real text and line art.

I upgraded to Acrobat 9 Pro Extended.  However, the .pdf to .doc/.rtf is still quite terrible under Acrobat --at least for me.  I create very good pdf documents from groff, but many publishers now require .doc submissions.  Acrobat (to be fair, as well as just about every other tool I've tried -- but those aren't as expensive :)!) do not do a good job at all.  Equations (simple Greek fonts?) and placement and fonts are not at all well converted.  A sample example is
http://ingber.com/smni06_ppi.pdf .

Lester

Last edited by ingber (2008-10-28 10:44:10)