I am trying to OCR a book using tesseract however I do not want to type each individual page as it is a book near 1000 pages. How would I accomplish this
I tried tesseract *.tif * which gives me this read_params_file: parameter not found: II*
Add a comment
|
To avoid overwriting the output file for every new page, I did:
for i in *tif; do b=`basename "$i" .tif`; tesseract "$i" "$b" hocr; done
It worked fine.
(I found this solution here.)
I tried this and it works
for i in *.tif ; do tesseract $i outtext; done;
Make sure when you are in terminal you change directory to the location of all of the tif files.
-
doesn't work does the same as <code>tesseract input output</code> I tried replacing outtext as * but that doesn't work either. Just renders that error – user140393 Mar 26 '13 at 11:22
-
Then reinstall tesseract as it worked for me unless you are using a really old version. – Meer Borg Mar 26 '13 at 21:30
-
I reformatted my linux OS and did an install of ubuntu. Grabbed synaptic package manager, then did a sudo apt-get update && sudo apt-get upgrade. Then rebooted. Then wen't on to install tesseract-ocr in synaptic. Works, but keeps overwriting the file for every new page. – user140393 Mar 27 '13 at 5:48
-
-
Does it even give you an error if you do it on one file? If so upload file and I can try to see if its the file thats causing the issue – Meer Borg Mar 27 '13 at 5:56
You can use VietOCR, which uses Tesseract and supports batch/bulk OCR.
I wanted convert Kannada (Indian language) as searchable pdf
I modified this
for i in *tif; do b=basename "$i" .tif; tesseract "$i" "$b" hocr; done
AS BELOW
for i in *tif; do b=basename "$i" .tif; tesseract -l kan "$i" "$b" pdf; done
It worked well for the entire folder was converted from image to searchable pdf Tnakyou
