How to recognize kinds of letters and digits?

[Borneq]

void TextWindow::recognize(const char *imagepath)

{

    Pix* pixs = pixRead(imagepath);

    if (!pixs)

    {

        fprintf(stderr, "Cannot open input file: %s\n", imagepath);

        exit(2);

    }

    tesseract::TessBaseAPI api;

    const char* lang = "pol";

    const char* datapath = "/usr/share/tesseract-ocr";

    tesseract::PageSegMode pagesegmode = tesseract::PSM_AUTO;

    tesseract::OcrEngineMode enginemode = tesseract::OEM_DEFAULT;

    int init_failed = api.Init(datapath, lang, enginemode);

    if (init_failed)

    {

        fprintf(stderr, "Could not initialize tesseract.\n");

        exit(1);

    }

    api.SetImage(pixs);

    tesseract::Orientation orientation;

    tesseract::WritingDirection direction;

    tesseract::TextlineOrder order;

    float deskew_angle;

    tesseract::PageIterator* it = api.AnalyseLayout();

    if (it) {

        it->Orientation(&orientation, &direction, &order, &deskew_angle);

        printf(

                    "Orientation: %d\nWritingDirection: %d\nTextlineOrder: %d\n"

                    "Deskew angle: %.4f\n",

                    orientation, direction, order, deskew_angle);

    } else {

        //ret_val = 1;

    }

    delete it;

    pixDestroy(&pixs);

    ////

    tesseract::PointerVector<tesseract::TessResultRenderer> renderers;

    PreloadRenderers(&api, &renderers, pagesegmode, "wynik");

    if (!renderers.empty()) {

        bool succeed = api.ProcessPages(imagepath, NULL, 0, renderers[0]);

        if (!succeed) {

            fprintf(stderr, "Error during processing.\n");

            exit(1);

        }

    }

}

Above procedure recognize text from image and save it to file "wynik.txt". But my images are letter addressing: first is first and last name WITHOUT digits, next is address and post code eq 12-345 - ONLY digits. How can I use this information for more accurate recognision of text?