I recently upgraded from DTP 1.5 to DTPO 2 specifically for the ability to OCR image-only PDF files. I just installed DTPO2 B3R2 and am trying out the “Convert to Searchable PDF” feature. So far, I’m not happy with the resulting file size and image quality I’m getting, and I wonder if there’s a way to do better.
The original PDF image scans are rather low resolution, 72 DPI. However their bit depth is only 1 bit. In other words, a pixel is either full black or full white with no shades of gray. The text is surprisingly (and quite comfortably) readable despite the low resolution. And the file size is quite small. A typical page is about 100 kBytes.
When I convert these files to searchable PDFs, the resulting file is much larger, and the text is not as readable. To get reasonably readable results, I end up with a file that’s about 8-10x the size of the original. When I zoom in close, I can see that there are now shades of gray around the letters.
So I’m wondering if there’s some way to set the bit depth of the conversion process to 1 bit. In my case, most of the documents don’t have photographic pictures in them, so I don’t really care about accurate gray scale (or color) conversion.
Any words of advice?