Problem with single line three character Chinese

[Sayang]

a)  Tesseract correctly OCR'd eight (>30 character) lines of Chinese, scanned from a book

b) Tesseract seemed to fail OCR'ing a single line image with three characters (xingqisi - Thursday)

   (i) Four different fonts were tried - so four different single line images - attached.
   (ii) The binary data produced by Tesseract for each of the four attempts was identical
   (iii) There were no error messages.

Any suggestions would be greatly appreciated.

[Paul]

Did you try the options -psm 7 or -psm 8?

Probably you will get better results by using one of them.

Paul

[HanmoLingfeng]

Sometimes  this problem in my project,and the  image(.jpg) file  convert into the file(.tif) which is empty . Do you know  the reason ?
Thx your help.