Abstract: The method may be used in the devices for banknote detecting, counting or sorting; it may be used to determine the banknote main characteristics: its currency type and denomination. The method is based on a computational processing of the banknote digital image formed in the device during scanning. Increase in the banknote processing speed is the technical result. The first variant of the method envisages a preliminary classification and further checking of correspondence to selected possible classes according to several criteria typical for each particular class. For checking of correspondence, an admissibility of the value of interrelation characteristic for two areas is validated. According to the second variant, there is made classification on the base of a banknote digital image and for each possible class the measure of the banknote correspondence to the image is calculated and used for ranking, the banknote is referred to the class closest to it by the measure of correspondence.

Abstract: The invention group relates to the banknote detecting means. The technical result is increase in the banknote identification accuracy. The method is based on a computational processing of the banknote scan formed in the device during scanning. The banknote digital image is separated into areas; for each area a function is calculated and a feature vector is composed with further calculation of the distance to the known classes represented by the parameters available beforehand. Moreover, the calculated function of the banknote digital image allows calculation of its signature in a form of binary number that is compared with the classes known beforehand.

Abstract: The method may be used in the devices for banknote detecting, counting or sorting; it may be used to determine the banknote main characteristics: its currency type and denomination. The method is based on a computational processing of the banknote digital image formed in the device during scanning. Increase in the banknote processing speed is the technical result. The first variant of the method envisages a preliminary classification and further checking of correspondence to selected possible classes according to several criteria typical for each particular class. For checking of correspondence, an admissibility of the value of interrelation characteristic for two areas is validated. According to the second variant, there is made classification on the base of a banknote digital image and for each possible class the measure of the banknote correspondence to the image is calculated and used for ranking, the banknote is referred to the class closest to it by the measure of correspondence.

Abstract: The invention group relates to the banknote detecting means. The technical result is increase in the banknote identification accuracy. The method is based on a computational processing of the banknote scan formed in the device during scanning. The banknote digital image is separated into areas; for each area a function is calculated and a feature vector is composed with further calculation of the distance to the known classes represented by the parameters available beforehand. Moreover, the calculated function of the banknote digital image allows calculation of its signature in a form of binary number that is compared with the classes known beforehand.