FIELD: physics.
SUBSTANCE: device is proposed for implementing a method for determining the possible division of a word image into symbol images to convert a document image into an electronic document. The device comprises one or more processors, one or more memory devices, and a program implemented as a set of digital instructions stored on one or more memory devices and executed by one or more processors. The program provides an image of a text string in one of the languages, the letters of which are not separated, when writing with spaces, and also converts the resulting image of a text string into one of the languages, the letters of which are not separated, when writing with spaces to a sequence of parameterized characters, where each parameterized symbol corresponds to one, two or more fragments of the text string in the image.
EFFECT: increasing the efficiency of optical recognition of the text symbols in different languages.
20 cl, 73 dwg
