FIELD: information technologies.
SUBSTANCE: invention relates to the recognition systems and methods using the artificial intelligence. Result is achieved by fact, that performing the hieroglyph image reception by the processing device, hieroglyph image supply as the input information to the machine learning trained model in order to determine the components combination on plurality of positions in the hieroglyph and the hieroglyph classification as the language specific symbol, based on the certain components combination on plurality of positions in the hieroglyph image. Another method may involve training the machine learning model to determine the components combination on plurality of positions.
EFFECT: technical result consists in the one or more machine learning models structures simplification and reduction in the amount of processing and computing resources necessary for the hieroglyphs recognition.
20 cl, 15 dwg
