OCR - optical character recognition, with MODI - microsoft office document imaging.

Microsoft has developed tool for OCR, which we can use it as a COM object.

First you need to install MODI on your computer, then, you can use it like this:

Document md = new Document();
String ocrText = String.Empty;

const bool ocrOrientImage = false;
const bool ocrStraightenImage = false;
md.OCR(MiLANGUAGES.miLANG_ENGLISH, ocrOrientImage, ocrStraightenImage);

var image = (MODI.Image)md.Images[0];
var layout = image.Layout;

foreach (Word word in layout.Words)
 if (ocrText.Length > 0)
   ocrText += " ";

  ocrText += word.Text;
textBox1.Text = ocrText;

Only problem is if you want to really release the object md, then you ave to use SaveAs method something like:


string path = Path.GetDirectoryName(openFileDialog1.FileName);

md.SaveAs(path + "\\deleteMe.tif", MODI.MiFILE_FORMAT.miFILE_FORMAT_DEFAULTVALUE, 

Taken from here, and from this source code.

My source you can download from here, and exe file can be found here.