Maori-English OCR Example in DjVu
A presentation by PlanetDjVu, March 11, 2003
Maori Religion and Mythology, by Edward Shorthand is a book that has been scanned and then presented in both PDF and DjVu image-only formats by Case Western Reserve University Library.
Click here to view the presentation page at Case Western Reserve University Library.
You will notice that the DjVu version is presented as a single 1,080 KB file, while the PDF version, which totals 12,225 KB in size, had to be broken down into 9 separate files for effective web delivery.
Both versions are image only, as there are very few, if any, software applications available that can perform OCR (Optical Character Recognition) in both Maori and English.
We demonstrate in this presentation that the JRAPublish application is capable of OCRing in both Maori and English, and is capable of performing this OCR for the DjVu format! We are sure that that JRAPublish alone can accomplish this!
OCR Capabilities of JRAPublish
JRAPublish supports OCR in an astonding total of 176 languages, in any combination, although for best results, not more than 3 languages should be OCRed at one time. Only two of the many supported languages, Maori and English, are needed for this example book.
JRAPublish is a commercial OCR and file conversion application that is available today, at www.searchpdf.com.
With a license to this software, you can make PDF files, but DjVu output is disabled, as the current licensing terms quoted by LizardTech for DjVu are, sadly, way out of our reach. We cannot even confirm that LizardTech still offers OCR for DjVu in any language. There is no longer any mention of an OCR capability for DjVu on their website, nor are there any examples.
Because we don't forsee a change in these licensing terms, at least while LizardTech is still in business and with the current management, we have renamed JRAPublish to "PDFPublish". Still, if you acquire a license to PDFPublish (for PDF), we may someday be able to enable DjVu output for you, and then you can enjoy the benefits of this highly-compressed format with OCR for searchable text in many international languages.
Other Multi-language OCR Examples
Visit the International Gallery of PlanetDjVu for other examples of OCR with multiple languages, including Hawaiian, Zulu, Greek, Icelandic, Finnish and Czech!