问题
I am using tesseract as the OCR engine for my ANPR application. I have trained tesseract 3.01v with the numberplate font. But I need to know:
- Which files should be included in the tessdata folder?
- Should I use the same tessdata folder where tesseract 3.01v is installed?
- I have trained with tesseract 3.01v and I am using tessnet2 in my code so will it be a problem?
Following is the code that I tried it with but it keeps exiting from the DoOcr() method.
List<tessnet2.Word> ocrText = new List<tessnet2.Word>();
tessnet2.Tesseract ocr = new tessnet2.Tesseract();
ocr.Init(@"C:\Program Files (x86)\Tesseract-OCR\tessdata", "eng", true);
ocrText = ocr.DoOCR(bmpGrayScale, new Rectangle(rect.X, rect.Y, rect.Width, rect.Height));
foreach (tessnet2.Word word in ocrText)
Console.WriteLine("{0} : {1}", word.Confidence, word.Text);
Does anyone have an idea as to whats wrong?
回答1:
"3.01 is not backwards compatible with 2.04. The data files are different."
http://code.google.com/p/tesseract-ocr/wiki/ReadMe
For .NET library compatible with 3.01, look at the project at http://code.google.com/p/tesseractdotnet/ or https://github.com/charlesw/tesseract-ocr-dotnet.
来源:https://stackoverflow.com/questions/11674288/what-files-should-be-included-in-the-tessdata-folder-after-training-tesseract