How to convert scanned PDF manuals into editable ones

[daisizhou]

Hallo everyone

This is a scanned version of the manual. I tried to use the conversion technology available on the Internet but was unable to convert it into an editable PDF document.
I tried to convert it to WORD, but failed.
If anyone has a better way, please tell me, thanks

https://we.tl/t-bZ1LVsFj99

[BeBuLamar]

Do you have Adobe Acrobat? If you do it can use OCR to convert the scanned image to text. You need to use some form of OCR software to convert it to text first.

[HighVoltage]

In the past, I had best results with ABBY FineReader OCR to PDF

https://pdf.abbyy.com/

[djsb]

I use NormCap to extract code listings from PDF files.

https://dynobo.github.io/normcap/

It works well if working on one section at a time.

I have not tried extracting anything more ambitious (for instance, pages from a PDF service manual) with other tools.

[daisizhou]

OK  . I'm done, thanks

[Old Printer]

It's not just a PDF, it's whats in the PDF. PDF is a file format that can contain ASCI text as well as bitmaps (scans). If it was ASCI text it would not have to be converted, so it is actually a scan of a printed page you are after. OCR software has come along way in the last 25 years, but it is still at the mercy of the quality of the scan. If the printed page being scanned is a little fuzzy or the scan is not done at high resolution the software can only do so much. Capitol "B" for instance can easily be misread as an "8", and so on. If the subject is anything technical a world class proofreader has to examine the conversion letter by letter to be sure it is accurate. Often it is just more efficient in the long run to keyboard it from scratch.

[calzap]

A good, but cheap, pdf editor is PDF-Exchange Editor.  Much cheaper than Acrobat and has worked well for me … although I only do simple tasks like adding/removing text and images from existing documents.  The basic version costs less than US$6 per year, advanced version less than US$8 per year.  There is a free version too, but supposedly it will leave a watermark on documents created with it.  I’ve never seen a watermark appear on existing docs I’ve edited with the free version.

Anything that is scanned will appear initially in the scanner in bitmap format and that is what is usually transmitted to the computer, often as a series of bytes rather than a complete image.  Image processing software can be used to do minor edits like erase text or add text to open spaces.  Many image processing programs can create PDF files from a group of images, and there is software to combine those PDF files into a single PDF document.  However, the resulting PDF file will just be a series of images and more difficult to edit than the original bitmap files.  To have a fully editable version, the original images will have to be run through an OCR program.  All OCR programs can create plain text files.  Some may create Word or PDF files.  Be prepared to spend some time editing the OCR output.

If you have an existing PDF file that was created by scanning printed pages, first it will need to be burst to single page PDFs.  Then those will have to be run through an OCR program, perhaps needing to be reconverted to an image format first.  Then the OCR output needs to be edited by a human before it can go into a combined PDF file.  Maybe someone knows of an OCR program that take a PDF consisting of bitmaps directly to a PDF consisting of editable text (i.e. vector format).

Mike