0

Goal: book -> XML which contains text and structure information (such a section hierarchy and heading information)

I'm not sure such software even exists. Theoretically the information is there: font-size/special ornaments would designate section information, multiple columns should be parseable, but I'm not aware of any such software.

  • any OS
  • any price
  • no need for batch processing
  • no need for the software to perform the scan process
  • book can be in any format (I'll make it work which whichever format is possible)
2

If you would like to work with a commercial product then you should check out the LEADTOOLS OCR SDK. With this, you'll be able to load in your book while running OCR all of the text and then saving your output to XML. Your XML output will consist of the confidence level of the recognition, the bounds of the text and what each element is. You can do this just a few lines of codes:

// Initialize the OCR engine
ocrEngine = OcrEngineManager.CreateEngine(OcrEngineType.Advantage, false);

//Startup the engine
ocrEngine.Startup(null, null, null, @"C:\LEADTOOLS 19\Bin\Common\OcrAdvantageRuntime");

//Load image
RasterImage image = ocrEngine.RasterCodecsInstance.Load(file, 1, CodecsLoadByteOrder.Rgb, 1, -1);

//Add pages
ocrDocument.Pages.AddPages(image, 1, -1, null);

//Recognize Page
ocrDocument.Pages.Recognize(null);

//Save Text
ocrDocument.SaveXml(OcrXmlOutputOptions.None);

Disclaimer: I am an employee of this product

  • I don't see how this would retain structure information such as section information in the xml. Moreover, it seems as if the LEADTOOLS product doesn't add anything to my use case that OPENCV + Teseract would minus the $3000 price tag. – Joe Jan 23 '17 at 3:35

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.