How can I turn a multi-page TIF file into a PDF? [duplicate]

I have a large multi-page TIF image file.

I scanned in a book a few days ago and I accidentally scanned it as a multi-page TIF. I really want to turn it into a more accessible PDF file.

Do you have any idea how this can be done?

Solution 1:

If you are fine with just converting the file format (without making the text searchable (via OCR as pointed out in the answer by Jonathan Ben-Avraham), this will be a perfect task for convert from the ImageMagick package (available for various platforms: Win, Linux, Mac). When you install, be sure to select the option to include the "convert" legacy utility.

Simply use that command to convert your multi-page TIFF file

convert multi-page.tif book.pdf

However, you should employ the -compress Parameter. If you have a 24bit colored TIF, you can use LZW (lossless) or JPEG (lossy), e.g.

convert -compress LZW multi-page.tif book.pdf

If your TIFF is black and white (please make sure, it is really saved with only 1bit colordepth), I usually use -compress Fax.

Here is an example for a 5 page sample file, with some text in it -- the actual ratio between the compression algorithms depends of course on the content of your file:

24bit JPEG:  1294kB
24bit LZW:   1759kB
1bit  Fax:    135kB

After that conversion you still can run a OCR tool (like Adobe Acrobat Pro) to make the text search- and copyable.

Solution 2:

TIFF (Tagged Image File Format) is a pixel-oriented format that is intended for images. In order to get just the character data from a TIFF file you would have to use an OCR (Optical Chararacter Recognition) program. All OCR programs have some error rate. The format of the pages (header, footers, section headings, etc) also affect the ability of OCR to accurately guess the characters.

The language and particularly the writing system and font also affect OCR accuracy. If the book has an unusual non-western font with ligatures then the chance of getting goot OCR output are close to zero.

If the book is not primarily text, for example, a book with many screen shots or other images that are important, then OCR will not help you.

There are some Open Source OCR programs available, depending on the OS you are using. Google Drive has a free OCR service. WMMV.

Assuming that the original book you scanned is in paper format, then TIFF is about the best you can do in any event. You can make the document size more manageable by compressing the TIFF file. Try JPEG compression with various levels of compression to optimize document size against readability.