Tesseract OCR?
 

I've installed Tesseract OCR including Tesseract-gui from the Sourceforge tar packet.
I also installed tesseract3, Tesseract EQU and Tesseract OSD from the Synaptic repositories. (Did I do the right thing?)
After installation I rebooted and ran the gui which opened flawlessly, unfortunately there are no instructions or tutorial,
and so it's fairly guesswork and trial by error on my part. I watched an old tutorial on YouTube which said the file to recognise
has to be a .tiff file. I followed the instructions carefully and all appeared to go well as the new text file popped up on my desktop as planned.
The problem is there was no script ... it was blank.
I attempted the procedure differently several times except from the Terminal (I don't feel confident using the terminal yet) and
I still couldn't get it to work.
Perhaps I have a necessary file missing or I've confused the program by adding a segment that isn't necessary, I don't know!
I would love to get this program working as it would be a great help to me in my work.
So! if anyone is using this program and would like to help me set it up successfully I would be very thankful.



gael.

After a little playing around with the GUI for a while and installing Tesseract-OCR-ALL from Synaptic Repositories, the program now works.
I've also discovered that it will work with .jpeg and .png files ... it could work with others that I haven't tried.
If the writing is over the top of a picture, Tesseract gets some of it wrong. it is very good when the writing is on a plain white background as it gets
about 95+ correct.