Awesome help!
I was only getting garbage from the GetXText() methods. The problem was I
was using opencv matrices to load my images. I wasn't aware of how to read
in a Pix image until I saw the code snippet in this thread. Now my data
actually makes sense =)
On Thursday, December 15, 2011 5:03:47 PM UTC-5, braza wrote:
>
> Hi,
>
> The world of open source welcomes me with insufficient info/examples/
> documentation but with opened doors to ask ;)
>
> I`m trying just to recognize really clear and simple line of text in
> English like "Tess TEST 123.4 $15"
>
> now I have:
>
> //Tesseract block start
> CTessOCR *tess = CTessOCR::Instance();
> tess->api->SetVariable ("tessedit_char_whitelist",
> "0123456789");
> tess->api->SetVariable ("classify_bln_numeric_mode",
> "1");
> tess->api->Init ("./../../tessdata",
> "eng");
> #ifdef DEBUG_MODE
> tess->api->SetVariable ("tessedit_adaption_debug",
> "T");
> tess->api->SetVariable ("tessedit_draw_outwords",
> "T");
> tess->api->SetVariable ("tessedit_dump_choices",
> "T");
> tess->api->SetVariable ("tessedit_dump_choices",
> "T");
> tess->api->SetVariable ("interactive_mode", "T");
> tess->api->SetVariable ("tessedit_create_hocr",
> "T");
> #endif
> tess->api->SetVariable ("tessedit_single_match",
> "0");
> tess->api->SetVariable ("tessedit_zero_rejection",
> "T");
> tess->api->SetVariable ("tessedit_minimal_rejection",
> "F");
> tess->api->SetVariable ("tessedit_write_rep_codes",
> "F");
> tess->api->SetVariable
> ("tessedit_resegment_from_boxes",
> "T");
> tess->api->SetVariable ("tessedit_train_from_boxes",
> "T");
> tess->api->SetVariable ("textord_fast_pitch_test",
> "T");
> tess->api->SetVariable ("textord_no_rejects", "T");
> tess->api->SetVariable ("edges_children_fix", "F");
> tess->api->SetVariable ("edges_childarea", "0.65");
> tess->api->SetVariable ("edges_boxarea", "0.9");
> tess->api->SetVariable ("il1_adaption_test", "1");
> tess->api->SetPageSegMode (tesseract::PSM_SINGLE_LINE);
>
> Mat img = imread( "../../tess.jpg", CV_LOAD_IMAGE_GRAYSCALE );
> //err
> now
> tess->api->SetImage(convert_mat_to_pix(img));
> std::string text = tess->api->GetUTF8Text();
>
> It all fails in
>
> > match.exe!OpenBoxFile(const STRING & fname)
> match.exe!tesseract::Tesseract::ApplyBoxes(const STRING & fname,
> bool find_segmentation, BLOCK_LIST * block_list)
> match.exe!tesseract::TessBaseAPI::Recognize(ETEXT_DESC * monitor)
> match.exe!tesseract::TessBaseAPI::GetUTF8Text()
>
> Obviously it fails because I`ve never set input file name with boxes.
> But why would I need it? I already have trained data downloaded and
> put in tessdata: eng.traineddata, eng.cube.size etc
>
>
--
--
You received this message because you are subscribed to the Google
Groups "tesseract-ocr" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/tesseract-ocr?hl=en
---
You received this message because you are subscribed to the Google Groups
"tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/groups/opt_out.
