I have such a problem: in some cases, my code recognizes 9 as S, and in some cases everything is fine. I tried all the fixing methods: changed the config, and also completely optimized the picture in order for tesseract to determine the text with great accuracy, but apparently this did not help. Do you have any ideas about this?
This is my custom config: config = '--oem 1 --psm 13'
Below I have attached pictures in which the tesseract recognizes 9 as S, and when it recognizes 9 as 9
https://i.stack.imgur.com/xKyEi.jpg - 1
https://i.stack.imgur.com/mfOVw.jpg - 2
In the first picture, it recognized the text as following: WSNU-AJ6T-4M2M
In the second picture, it recognized the text as following: VAPI-29UI-WAPI