编译下面的程序操作系统必须在安装了tesseract库和leptonica库才可以

Basic example

c++ code:

#include <tesseract/baseapi.h>
#include <leptonica/allheaders.h>

int main()
{
    char *outText;

    tesseract::TessBaseAPI *api = new tesseract::TessBaseAPI();
    // Initialize tesseract-ocr with English, without specifying tessdata path
    if (api->Init(NULL, "eng")) {
        fprintf(stderr, "Could not initialize tesseract.\n");
        exit(1);
    }

    // Open input image with leptonica library
    Pix *image = pixRead("/usr/src/tesseract-3.02/phototest.tif");
    api->SetImage(image);
    // Get OCR result
    outText = api->GetUTF8Text();
    printf("OCR output:\n%s", outText);

    // Destroy used object and release memory
    api->End();
    delete [] outText;
    pixDestroy(&image);

    return 0;
}

GetComponentImages example

  Pix *image = pixRead("/usr/src/tesseract-3.02/phototest.tif");
  tesseract::TessBaseAPI *api = new tesseract::TessBaseAPI();
  api->Init(NULL, "eng");
  api->SetImage(image);
  Boxa* boxes = api->GetComponentImages(tesseract::RIL_TEXTLINE, true, NULL, NULL);
  printf("Found %d textline image components.\n", boxes->n);
  for (int i = 0; i < boxes->n; i++) {
    BOX* box = boxaGetBox(boxes, i, L_CLONE);
    api->SetRectangle(box->x, box->y, box->w, box->h);
    char* ocrResult = api->GetUTF8Text();
    int conf = api->MeanTextConf();
    fprintf(stdout, "Box[%d]: x=%d, y=%d, w=%d, h=%d, confidence: %d, text: %s",
                    i, box->x, box->y, box->w, box->h, conf, ocrResult);
  }

Result iterator example

There is posibility to get confidence value and BoundingBox per word from ResultIterator:

  Pix *image = pixRead("/usr/src/tesseract-3.02/phototest.tif");
  tesseract::TessBaseAPI *api = new tesseract::TessBaseAPI();
  api->Init(NULL, "eng");
  api->SetImage(image);
  api->Recognize(0);
  tesseract::ResultIterator* ri = api->GetIterator();
  tesseract::PageIteratorLevel level = tesseract::RIL_WORD;
  if (ri != 0) {
    do {
      const char* word = ri->GetUTF8Text(level);
      float conf = ri->Confidence(level);
      int x1, y1, x2, y2;
      ri->BoundingBox(level, &x1, &y1, &x2, &y2);
      printf("word: '%s';  \tconf: %.2f; BoundingBox: %d,%d,%d,%d;\n",
               word, conf, x1, y1, x2, y2);
      delete[] word;
    } while (ri->Next(level));
  }

Orientation and script detection (OSD) example

  const char* inputfile = "/usr/src/tesseract-3.02/eurotext.tif";
  tesseract::Orientation orientation;
  tesseract::WritingDirection direction;
  tesseract::TextlineOrder order;
  float deskew_angle;

  PIX *image = pixRead(inputfile);
  tesseract::TessBaseAPI *api = new tesseract::TessBaseAPI();
  api->Init("/usr/src/tesseract-3.02/", "eng");
  api->SetPageSegMode(tesseract::PSM_AUTO_OSD);
  api->SetImage(image);
  api->Recognize(0);

  tesseract::PageIterator* it =  api->AnalyseLayout();
  it->Orientation(&orientation, &direction, &order, &deskew_angle);
  printf("Orientation: %d;\nWritingDirection: %d\nTextlineOrder: %d\n" \
         "Deskew angle: %.4f\n",
         orientation, direction, order, deskew_angle);

Example of iterator over the classifier choices for a single symbol

  Pix *image = pixRead("/usr/src/tesseract-3.02/phototest.tif");
  tesseract::TessBaseAPI *api = new tesseract::TessBaseAPI();
  api->Init(NULL, "eng");
  api->SetImage(image);
  api->SetVariable("save_blob_choices", "T");
  api->SetRectangle(37, 228, 548, 31);
  api->Recognize(NULL);

  tesseract::ResultIterator* ri = api->GetIterator();
  tesseract::PageIteratorLevel level = tesseract::RIL_SYMBOL;
  if(ri != 0) {
      do {
          const char* symbol = ri->GetUTF8Text(level);
          float conf = ri->Confidence(level);
          if(symbol != 0) {
              printf("symbol %s, conf: %f", symbol, conf);
              bool indent = false;
              tesseract::ChoiceIterator ci(*ri);
              do {
                  if (indent) printf("\t\t ");
                  printf("\t- ");
                  const char* choice = ci.GetUTF8Text();
                  printf("%s conf: %f\n", choice, ci.Confidence());
                  indent = true;
              } while(ci.Next());
          }
          printf("---------------------------------------------\n");
          delete[] symbol;
      } while((ri->Next(level)));
  }

在linux下面编译C++ 程序的方法：

g++ -o myprogram myprogram.cpp -llept -ltesseract

如果你安装tesssract的时候不是通常的路径，需要手动指定头文件的include和库lib的路径。加上选项-I和-L如：

g++ -o myprogram myprogram.cpp -I/home/nick/local/include/tesseract -L/home/nick/local/lib -llept -ltesseract


c语言程序的例子：

#include <stdio.h>
#include <allheaders.h>
#include <capi.h>

void die(const char *errstr) {
        fputs(errstr, stderr);
        exit(1);
}

int main(int argc, char *argv[]) {
        TessBaseAPI *handle;
        PIX *img;
        char *text;

        if((img = pixRead("img.png")) == NULL)
                die("Error reading image\n");

        handle = TessBaseAPICreate();
        if(TessBaseAPIInit3(handle, NULL, "eng") != 0)
                die("Error initialising tesseract\n");

        TessBaseAPISetImage2(handle, img);
        if(TessBaseAPIRecognize(handle, NULL) != 0)
                die("Error in Tesseract recognition\n");

        if((text = TessBaseAPIGetUTF8Text(handle)) == NULL)
                die("Error getting text\n");

        fputs(text, stdout);

        TessDeleteText(text);
        TessBaseAPIEnd(handle);
        TessBaseAPIDelete(handle);
        pixDestroy(&img);

        return 0;
}

在linux下面编译C语言程序的方法和c++的方法是一样的，只不过换个编译器就好：

gcc -o myprogram myprogram.cpp -llept -ltesseract

如果你安装tesssract的时候不是通常的路径，需要手动指定头文件的include和库lib的路径。加上选项-I和-L如：

gcc -o myprogram myprogram.cpp -I/home/nick/local/include/tesseract -L/home/nick/local/lib -llept -ltesseract

运行方法：
把一个图片复制到当前目录下命名为 img.png

./myprogram img.png