Exit
  • Global community
    • Language:
      • Deutsch
      • English
      • Español
      • Français
      • Português
  • 日本語コミュニティ
  • 한국 커뮤니티
0

Rendering Vector Font from Bitmap Scan

New Here ,
Jul 22, 2019 Jul 22, 2019

I scanned a document and the result was a bitmap rendering of the font (an image of the document) which is, by its nature, slightly blurry.

I used the "Enhance" feature of Acrobat to try and sharpen the text, as well as other enhancements.

The result was a document that had a vector font overlayed on the old bitmap font.  The text and formatting of the new vector font was perfect; it remained the same as the underlying scan with the exception of the height of the font (which did not effect the overall format).

The problem is that the old underlying bitmap text remains in the document, so it looks like the page was fed through a printer twice, first with the scanned image, and then with the new font.

How can I remove the underlying image text (bitmap) and only retain the new vector font?

TOPICS
Scan documents and OCR
767
Translate
Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Expert ,
Jul 22, 2019 Jul 22, 2019

Hi qtpress,

When you scan a document, you DO get a bitmapped image so that is normal, that's what's supposed to happen until you OCR the result of the scan. The quality of the scan is based on the resolution that you scanned the image at AND the quality of the original document. That is, if you scan a blurry document you will get a blurry scanned document. There is no way to sharpen a blurry scan. There is a feature letting you "sharpen text" but that sharpening is no different than sharpening in Photoshop: it enhances contrast (to make dark things darker and light things lighter) but it does not actually sharpen.

Another thing that can help your scan is to save it as a tif image, not jpg. Jpg is a lossy format that affects the quality of the document by intentionally losing data. This can be acceptable on photographic images but will never be acceptable on things like text (unless you set the compression to zero and even then it's not advised).

Now, to your question, when you set your OCR settings, there are two (or three if you are using an older version) as shown in this screenshot.

2019-07-22_14-00-40.png

From Adobe:

Searchable Image
Ensures that text is searchable and selectable. This option keeps the original image, deskews it as needed, and places an invisible text layer over it. The selection for Downsample Images in this same dialog box determines whether the image is downsampled and to what extent.

and

Editable Text & Images
Synthesizes a new custom font that closely approximates the original, and preserves the page background using a low-resolution copy.

Because I haven't seen the page in question, your scanning of that page, and a bunch of other things, I hope you can get enough information from what I've written and submitted to come to a better result.

Please let us know if any of this makes sense and helps you.

Translate
Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
New Here ,
Jul 22, 2019 Jul 22, 2019

Thanks for the help,

To be clear, my goal is to have acrobat render the new font and replace the bitmap font.

My attempt worked in that a new vector font was rendered, but the new font was "on top of" the old bitmap scanned font on the same document.

I would like to keep the new font and get rid of the old. I just can'tseem to figure out how to do this.

Translate
Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Expert ,
Jul 22, 2019 Jul 22, 2019

Hi QTPress,

Than you want to select the "Editable Text & Images" option.

But please if your original scan is not good, you are only going to be spending extra time time to bypass the problems. If your scan is fine (and the original copy is poor), you're doing as best as you can under the circumstances.

Please let us know how this works out.

Translate
Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
New Here ,
Jul 22, 2019 Jul 22, 2019
LATEST

The original document and scan arent really that bad. I'm just trying to convert the bitmap text into vector text, which it looks like acrobat cand do (and does pretty well) except that it renders the new vector over the old bitmap.

I'll try your suggestion though.

Translate
Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines