FIELD: image processing means.
SUBSTANCE: invention relates to processing images of scanned documents and other images containing text. Technical result is achieved by identification of image symbols in the scanned document image containing text; for each page of the document, for each image of the symbol on the page, for identifying each grapheme from the set of graphemes, which corresponds to the normalized image of the symbol relative to the symbol standard from the set of symbol standards, sorting the identified graphemes by the frequency, with which the identified graphemes correspond to the normalized image of the symbol relative to the symbol standards in the set of symbol standards, and using the sorted identified graphemes to select the symbol code that represents the normalized image of the symbol; and preparing a processed document comprising character codes that represent normalized symbol images from the scanned image of the document, and storing the processed document on one or more than one from one or more than one memory devices and memory modules.
EFFECT: technical result is to increase the efficiency of recognition of optical symbols.
20 cl, 52 dwg
