Recognize random characters with tesseract

[1609Tommi]

Hello,

so I am a total newbie with tesseract and coding in general, have fooled around with tesseract for a couple of hours now, and just can`t get acceptable outputs.
The context is that i have several pages of random characters printed on paper, or now as .pngs, and want to get them back in a txt. file. How did this happen? Well, 3 years ago i invested a (very)
small amount in bitcoins, but as the price was rather stable and it wasn`t much money it all, i never sold them, and after a while forgot about them. And stupid me also reseted my pc :(. But, i printed my wallet.dat, so here I am.
Maybe you can help me.
The attached png is the picture with which i tried it, and the txt is my result.

So is there a way to improve the character recognition?
Or do i have to type it :D

Thank you,

Thomas

[adarsh shukla]

Tried to run tesseract with your image seite1_1.PNG and got same results as shown by you. But it seems it will resolve the issue if you use proper PSM mode.
for example: 


tesseract -psm 6 /home/adarsh/Downloads/seite1_1.PNG out
 This will give you this output:
AAAAAAEAAAAAAAAAYj EFAAkAAAAAIAAAAAkAAAAAAAAIAAAAAAAAAAAAAAAAAAAAIAAAAAQUAAAA
AJQAquGJgAMAAAMMAAAMAIAAAAAAAAAIAAAAAEAAAAMAAAMAMMMAMAAANMAAAMA
AAAMAAAAAAAAAAANMAAAAMMNMMMAAAAAAMMAAMMAAAAAAAAAAAAAANMAAAAAAMA


This is correct to single words as i checked.
 Happy Coding.