why can't it read the single character on the picture?

[umutozgur]

why can't it read the single character on the picture?

[ivanstepanovftw]

Try this:

TESSDATA_PREFIX = "/usr/share/tesseract/tessdata"
tesserocr_languages = ["eng", "ara"]
api = PyTessBaseAPI(path=TESSDATA_PREFIX, lang="+".join(tesserocr_languages))

api.SetImageBytes(
    imagedata=pixmap.samples,
    width=pixmap.w,
    height=pixmap.h,
    bytes_per_pixel=bpp,
    bytes_per_line=pixmap.stride,
)
api.SetPageSegMode(tesserocr.PSM.SINGLE_CHAR)  # <- important
api.Recognize()
ocr_text = api.GetUTF8Text()

[zdenop]

without providing an input image you are alone with your problem...

[ivanstepanovftw]

Sure!

I think this issue should be closed as resolved.

If image contains single character, read this:

Tesseract works best on images which have a DPI of at least 300 dpi, so it may be beneficial to resize images

Here is also plot for you: