Method for automatic reading of addresses in more than one language of which at least one language has non-Latin characters using an OCR device and address analysis unit

摘要

Method has the following steps: reading of the characters in the areas of the address blocks using OCR unit for the represented languages, and presentation of the results in a transliteration representation (13); determination and classification of the different address elements from the OCR units together with language related syntax rules (14); and verification (15) of the identified address elements using an address database with different language dependent transliteration versions for each database entry. An Independent claim is made for a corresponding device for automatic reading of addresses with an OCR unit and an address analysis unit.