Qualification of Tesseract dll interface - inquiry regarding a feature.

[TekGnosis]

My apologies for not being as familiar with finding out the answer
myself from the code, but I am not much of a programmer despite
needing to be be sometimes.

Note that I have not found exactly what I need to know via the
'search' function.

I have need to examine the screen in coordinates that I can pass from
the interface I am using (autohotkey if it matters) and reflect the
result to a memory address in ASCII. Is this possible without
resorting to a stage where the section of screen is dumped to a file
for examination? I am finding that GOCR is not able to do this and
found it frustrating. This application is fullscreen 3d and the
continued disc churn for constant checking would be Bad.

Alternately, I can extract the pixels using another application and
send a pointer to the memory address containing the text to be OCR
after 'cleaning it up' if necessary.

**
So in short I need to pass screen coords (or memory pointer to the
bitmap) to the dll and retrieve back a pointer to an address with the
OCR results.
**

Either way, I very much need to know if it is at least Possible so I
can be sure learning this applications interface is worthwhile.
Otherwise I can use SimpleOCR or Textract, but they are not GPL and I
don't like that...

Thanks much for your help in advance.

Nathan

[Scan...@gmail.com]

If you can get the screen bitmap into memory, you could then possible
pass to the dll and get the character structure back that tess makes.

You would need to know the dimensions and depth of the screen bitmap
and need to convert the returned structure into something you can use.