OCR: How to use OCR result from other OCR engine?
In This Topic
If you have OCR result in hOCR format, you can import hOCR result - please read how to do this
here.
If you do not have OCR result in hOCR format, you can create OCR result programmatically. For doing this you need to do the following steps:
- Create OCR document (instance of OcrDocument class)
-
For each recognized page:
- Create OCR page (instance of OcrPage class) and add created page to the page collection (OcrDocument.Pages property) of OCR document.
-
For each region, which is recognized on page:
- Create OCR region (instance of OcrTextRegion class) and add created region to the region collection (OcrPage.Regions property) of OCR page.
-
For each paragraph, which is recognized in region:
- Create OCR paragraph (instance of OcrParagraph class) and add created paragraph to the paragraph collection (OcrTextRegion.Paragraphs property) of OCR region.
-
For each text line, which is recognized in paragraph:
- Create OCR text line (instance of OcrTextLine class) and add created text line to the text line collection (OcrParagraph.TextLines property) of OCR paragraph.
-
For each word, which is recognized in text line:
- Create OCR word (instance of OcrWord class) and add created word to the word collection (OcrTextLine.Words property) of OCR text line.
-
For each text symbol, which is recognized in word:
- Create OCR text symbol (instance of OcrSymbol class) and add created text symbol to the text symbol collection (OcrWord.Symbols property) of OCR word.