News | April 26, 2011

Tips For Recognizing Documents With Mixed Languages

logo.jpg

With the increasingly global nature of business, finding a perfect solution to accurately process multilingual documents is critical.

FineReader is designed to process a wide range of multilingual documents with unmatched accuracy and layout retention. FineReader can recognize documents containing any language combination—even those with European, Cyrillic and Asian languages. Whether you want to convert an overseas contract or digitize a foreign newspaper, check out these tips for the best OCR results possible.

FineReader features automatic language detection for supported European languages. If your document contains mixed languages that includes Asian languages, we recommend you disable the automatic analysis and OCR option first and then follow the instructions below.

  1. Select More languages... in the Pages window from the Document Languages list. Select Specify languages manually from the Language Editor dialog box and select needed languages (e.g., English, Spanish and Chinese) from the language list.
  2. Scan or open images after disabling Detect page orientation. The dual page splitting option should be used only if all page images have the correct orientation. The pages will be added to the current ABBYY FineReader document after the command is executed.

    Important! When scanning, be sure that the pages are properly centered on the scanner's glass plate. If the skew is too large, the text may be recognized incorrectly.

  3. To draw areas on the image manually, use the tools under Image window to adjust area shapes and area borders.

    Note. If the structure of your document is simple, you can launch automatic layout analysis. Click the Analyze button or press Ctrl+E on the toolbar of the Image window.

  4. If there are areas on the image where text is written in only one language, select these areas, then select the language of the text area (e.g., Spanish) on the Area Properties panel under the Image window.

    Important! You can only specify a language for areas of the same type. If you select both text and table areas, you won't be able to specify a language.

  5. Click Recognize.

Follow these simple steps and you'll have the most accurate OCR possible for your multilingual documents.

To learn more about FineReader visit our website (http://finereader.abbyy.com/).

SOURCE: ABBYY