PDF to Text using OCR

Hello!

I can not understand how to use the SDK. Need to recognize PDF and get the entire text, as well as its position and size (width and height, or top-left and bottom-right coordinates).

Now I've settled on this:

            IOcrEngine ocrEngine = OcrEngineManager.CreateEngine(OcrEngineType.Professional, false);
            ocrEngine.Startup(null, null, null, null);

            ocrEngine.AutoRecognizeManager.EnableTrace = false;
            ocrEngine.AutoRecognizeManager.MaximumThreadsPerJob = 0;
            ocrEngine.AutoRecognizeManager.MaximumPagesBeforeLtd = 0;
            ocrEngine.AutoRecognizeManager.JobErrorMode = OcrAutoRecognizeManagerJobErrorMode.Abort;
            ocrEngine.AutoRecognizeManager.PreprocessPageCommands.Clear();

            OcrAutoRecognizeJobData ocrJobData = new OcrAutoRecognizeJobData(path,  DocumentFormat.Html, "output.html");

            IOcrAutoRecognizeJob ocrJob = ocrEngine.AutoRecognizeManager.CreateJob(ocrJobData);
            ocrEngine.AutoRecognizeManager.RunJob(ocrJob);

As a result, getting an exception: PDF codec is needed to use this feature.

We plan to use v17 SDK, in .net 4 project.

Could you please explain to me what I'm doing wrong. Could you please explain to me what I'm doing wrong.

Thanks in advance.

To resolve the problem, you might need to add the Leadtools.Codecs.Pdf.dll file as reference in your project.

However, to get the recognized words locations, you might need to use the low-level OCR methods, such as ocrDocument.Pages.AddPage() and ocrPage.Recognize(). And then you can get the recognized character data of the OcrPage by using the IOcrPage.GetRecognizedCharacters Method.
For more information and sample code, please see the help topic "GetRecognizedCharacters Method" in the LEADTOOLS .Net documentation.

Thanks,
Maen Badwan
LEADTOOLS Technical Support