Since OCR accuracy depends on OCR-ready fonts, how to list OCR-ready fonts for Adobe Acrobat Pro?

[Sunny26326757tb7e]

Since OCR accuracy depends on OCR-ready fonts, how to add know if a Font is being used for Recognize Text. 

I am experiencing inaccurate Text Recognition from Scan and OCR of Images given to me.

• Use OCR-ready fonts. When creating digital documents, use standard, machine-readable fonts like Arial, Times New Roman, or Calibri. Avoid decorative or handwritten fonts, as OCR may struggle to interpret them accurately.
• Check language settings. Ensure that the OCR software is set to the appropriate language for the scanned text. OCR engines perform best when configured for the specific language of the document.

Use OCR-ready fonts; Avoid decorative or handwritten fonts, as OCR may struggle to interpret them ac... 

[gary_sc]

Hi, @Sunny26326757tb7e, Thank you for pointing to this Adobe article; I was unfamiliar with it.

It took me a while to understand what they meant when they talked about "OCR-friendly" fonts, as that was new to me. I've been using OCR technology with a variety of OCR software for about 25-30 years and I've never EVER heard of this issue. As much as I could figure out from the information presented, it has to do with the fonts USED in the document and has NOTHING to do with fonts in your system. When you think about it, if you did not have Times New Roman on your computer, to think that that would affect how successful you are in scanning a serif font is, well, very strange.

Font recognition success is determined by how well the company that creates the OCR software trains it. Script is the most challenging becuase of the various nuances within any script. Can it be done? Sure, but OCR software companies probably spend more time on the challenges with basic fonts as it is.

Simply, what that linked document is saying here is that OCR success is partially determined by selecting documents to scan that have OCR-friendly fonts in the document to be scanned. Or rather, if you have (say) written or scripted text in a document, do not expect to have tremendous success in OCR — and there's not much you can do about that.

On a side note, I was very amused that this document referred to "dpi" when scanning. That is sadly wrong as dpi (dots per inch) refers to printing. Ppi (pixels per inch) refers to scanning and image resolution in digital software. Small point, but always amusing. 

One other item of note: Adobe Acrobat cannot scan anything. Rather, it uses some software called "TWAIN" to link Acrobat to access your scanner's software. So, when you are scanning from within Acrobat, you are actually linking to the scanner's software. When you have issues during scanning, there are likely to be some glitches in this link. When and if that happens. you are probably best off to just go straight to the scanner's software and not bother with the middleman. It can be more trouble than it's worth. 

That article does correctly point the user to save there scanned documents into the PNG or TIFF format. Personally I strongly recommend that you use TIF if you scan using the scanner's software. It has to do with preferences built into Acrobat: if you open a PNG file into Acrobat, Acrobat will automatically convert the PNG to PDF, but then you then have to manually select the process to OCR that document. If you do the same thing with a TIFF file, Acrobat will automatically open as a PDF and automatically process the OCR aspect. This simply saves you a step.

To elaborate on techniques for scanning success, please take a look at this blog I wrote for Adobe a number of years ago. It might be helpful to you.

https://community.adobe.com/t5/adobe-community-professionals/scanning-clean-searchable-pdfs/m-p/4785...

Good luck!