Project Outcomes

• Theoretical results about construction of Weighted Levenshtein Automata for OCR correction.
• Large-size Bulgarian, Russian, German and English Electronic Dictionaries with OCR correction aiding data.
• Multilingual very-large size consolidated Bulgarian-Russian-German-English Dictionary with OCR correction aiding data.
• Probability tables of symbol-dependent recognition errors for commonly used Cyrillic and Latin Fonts.
• Word collocation table for Bulgarian for OCR correction.
• Robust and efficient software system for OCR correction based on the Levenshtein automata framework and the word collocation.