How to create a searchable PDF from a redacted document where the words redacted do not appear in th

The attached project is a Console application using C#, Visual Studio 2010, and LEADTOOLS .NET Library version 18.0.

The redacted document could be any of the LEADTOOLS supported formats using the LEADTOOLS Redaction Annotations.

The strategy that was used:

- Input: A file containing LEADTOOLS Annotations. The annotations can either be stored in a separate file or within a TIFF tag if the document is a TIFF document.
- The application attempts to load the annotations into memory from either location.
- Once read into memory, the application enumerates each annotation obtaining its location on the image.
- Using the location it fills in the area of the image under the annotation with a solid white rectangle. This essentially “erases” the word from the document.
- It runs this document through our OCR so all words are recognized minus the words that were redacted ("erased").
- Using the same function which erased the words, instead of filling in the rectangle with white, it fills the rectangle with black, thereby making it appear redacted, when in fact it has simply been drawn over.
- It then updates the image the OCR will use to save the results of the recognition with the image containing the black rectangles.
- Finally the results are saved to disk. The PDF will contain the searchable text (excluding the redacted words) and the image displayed will be the image with the black rectangles representing the redactions.