How to read text from scanned pdf

I have been using Tesseract in a C# program but it has been a while therefore I do not remember much. Also they apparently have made substantial improvements after I wrote my program.

Tesseract documentation | Tesseract OCR (tesseract-ocr.github.io)

When I posted this I did not see that this question is from half a year ago. I hope this helps someone at least.

I am unaware of any free API that can do what you are looking for, but you can use the LEADTOOLS Recognition SDK technology in your application.

https://www.leadtools.com/sdk/ocr

You can leverage the IOcrEngine class, to gather the text of any page of the scanned document. LEADTOOLS also offers the ability to use TWAIN to gather the image from the scan job to easily facilitate the scanned image to the IOcrEngine for recognition.

Here is some sample code:

string file = @"FILE PATH TO INPUT IMAGE FILE";
string outFile = @"FILE PATH TO OUTPUT PDF FILE";

// Initialize the OCR engine  
IOcrEngine _ocrEngine = OcrEngineManager.CreateEngine(OcrEngineType.LEAD); 
// Start up the engine  
_ocrEngine.Startup(null, null, null, @"\LEADTOOLS22\Bin\Common\OcrLEADRuntime"); 

// Create a document  
using (IOcrDocument ocrDocument = _ocrEngine.DocumentManager.CreateDocument(null, OcrCreateDocumentOptions.AutoDeleteFile)) 
using (RasterCodecs _codecs = new RasterCodecs())
{
    CodecsImageInfo _info = _codecs.GetInformation(file, true);
    
    if (_info.TotalPages <= 3)
    {
        for(int i = 1; i <=3; i++)
        {
            using (RasterImage _image = _codecs.Load(file, i))
            {
                // Create IOcrPage from loaded image 
                IOcrPage _ocrPage = _ocrEngine.CreatePage(_image, OcrImageSharingMode.AutoDispose); 
                // Recognize Text 
                _ocrPage.Recognize(null); 
                // Add the page  
                ocrDocument.Pages.Add(_ocrPage);
            }
        }
    }

    // Save as PDF 
    ocrDocument.Save(outFile, DocumentFormat.Pdf, null);
}

Hi Kavya,

I don't think any free source for this kind of major extraction and the concept is called "Optical Character Recognition".

You can refer to my article: https://www.c-sharpcorner.com/article/cognitive-services-optical-character-recognition-ocr-from-an-image-using-com/

Hi,

Use ITextSharp. 

Please refer below

https://www.codeproject.com/Questions/243295/Is-this-possible-to-Extract-Text-from-Scanned-PDF

Thanks,