Scan pdf file instead png

[Teo]

Is there an option to directly scan a pdf document containing multiple pages instead of the single png image?

[Essam Zaky]

What do you mean by "scan a pdf " ?

If you mean recognize pdf file , you can not recognize pdf file directly because it's unsupported format by leptonica

see the following read me

https://github.com/tesseract-ocr/tesseract/blob/master/doc/tesseract.1.asc

 

The workarround is to find a tool which can extract pdf to images , then write the extracted images  paths in one text file 

i.e. test.pdf will be

test.txt

     ../image/path/1.png

     ../image/path/2.png

     ../image/path/3.png

then call tesseract as follow

tesseract test.txt path/to/output -l eng 

the output.txt will contain all the recognition result for all files in test.txt

Best Regards

Essam

[Zdenko Podobny]

Tesseract is OCR images not documents (pdf, docx, odt etc..)

If you need multipage support use tif image format instead of pdf for scanning. 

Zdenko

so 28. 3. 2020 o 20:42 Essam Zaky <essa...@gmail.com> napísal(a):

--
You received this message because you are subscribed to the Google Groups "tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-oc...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/ffd9e7c7-8fdd-4ced-8707-eb6ceaf61b68%40googlegroups.com.

[Teo]

yes I meant just this. Ok thanks for your support.

[Teo]

Ok thanks