I am using Tesseract 4 and it's working fine but in some cases it is not performing well.
Example: Orignal Image

Image After apply Binarisation and 10 pixel border

And it's output is LOWEST PRICE OR IT'S with default psm
and by setting psm = 6 output is enya LOWEST PRICE OR IT'S
and by setting psm = 11 output is JOHNNY alate LOWEST PRICE OR IT’S
but still I am not able to get Janosik . And with psm 6,11 I also get wrong words. How can I improve the results.