The IOcrPage object used as the input for a page when OCR conversion is used.
This value is always null if OCR is not used in the conversion and is not available with all operations.
By the default, the engine will get the image from the current DocumentPage. You can override this behavior by supplying your own raster image object when DocumentConverterJobOperation.LoadRasterImage is called during OCR conversion.
Refer to DocumentConverterJobOperation for more information.