Method of text information recognition from a graphical file with use of dictionaries and other supplementary data

A computer-implemented method for text data recognition, comprising

• obtaining an image file from scanning device or from other source,
• preliminarily assigning at least a part of a list of applied supplementary data types and an order of application of the supplementary data types in the list, said list comprising, in order of application:
• a line-to-graphemes parsing information,
• a graphical element (grapheme) recognition quality,
• a whole words dictionary,
• a dictionary of permissible word fragments,
• rules, prescribed by applied standard data patterns or regular expressions,
• rules, prescribed by word disposition within the line or the paragraph,
• rules, prescribed by the document language peculiarities,
• rules, prescribed by the document type peculiarities, and
• supplementary rules for rare occasions,
• preliminarily assigning of an accuracy estimation for each type of supplementary data,
• performing one or more line-to-fragments parsing versions by reliably recognized spaces, said fragments presumably comprising single word images,
• building of line partition graph (hereinafter LPG) for each line fragment, said graph describing fragment-to-graphemes parsing versions, said graphemes presumably comprising character images,
• performing single graphemes recognition, using two or more classifiers of different types,
• assigning each said grapheme recognition version accuracy estimation,
• interpretation of grapheme recognition version as a character version,
• performing at least the following steps:
• the first step:
• for each LPG chain connecting initial node and final node, building a set of chains using all obtained recognized grapheme-to-character versions,
• calculating a total recognition accuracy level for each said chain,
• sorting obtained results in a total recognition accuracy descending order,
• the second step:
• analyzing all obtained character group versions using supplemental information about capital-small characters disposition,
• in a case of more than one grapheme-to-character recognition version being available, analyzing each said obtained recognition version with the application the supplemental data types in connection with the preliminarily assigned order to gain a prescribed recognition accuracy,
• assigning to each obtained version an accuracy estimation,
• discarding character versions having said accuracy estimation lower, than the preliminarily assigned level,
• sorting the remain versions in a descending order using pair wise comparison;
• the third step:
• performing a supplementary space recognition correction with respect to a previously mistakenly recognized spaces comprising:
• joining of previously mistakenly separated elements,
• separation of previously mistakenly combined elements.