Reputation: 337
I'm trying run tesseract-ocr over this image, unsuccessfully :
> wget https://i.sstatic.net/rXR44.png
...
> convert dOtlrvx.png dOtlrvx.tif
> tesseract dOtlrvx.tif out -psm 10 && cat out.txt
Tesseract Open Source OCR Engine v3.02 with Leptonica
Page 0
.
The recognized char is a dot "."
-psm 10 stands for "treat the image as a single character" so I think its the correct option to use. I also tried with other psm possible values, it does not work neither.
Anyone has an idea why is this not working ? Any suggestion is welcomed !
Thanks
Upvotes: 1
Views: 608
Reputation: 234
Create a new config file for tesseract, add this line tessedit_char_whitelist 0123456789
and then process your image: tesseract dOtlrvx.tif out -psm 10 your_config_file
.
This worked for me.
Upvotes: 1