Classification method and apparatus

Abstract

A method of deriving a classification for classifying items of currency into two or more classes comprises measuring known samples for each class, selecting a function corresponding to a non-linear mapping of the feature vector space to a second higher-dimensional space, mapping feature vectors to image vectors, and deriving coefficients representing N-1 axes, where N is the number of classes, inthe second space, obtaining values representing the projections of the image vectors for the measured samples onto the N-1 axes, and using those values to derive a separating function for separating the classes equivalent to a function in the second space.

Description

[0001] This application is a divisional application, the national application number of the parent case is 99815989.1, the filing date of the parent case is December 1, 199, the invention name of the parent case is "classification method and equipment", and the international application number of the parent case is PCT / IB99 / 02012. technical field [0002] The present invention relates to methods and devices for sorting items. The invention relates in particular to the sorting of coins or banknotes. Background technique [0003] Coins or banknotes inserted into mechanical devices, such as vending machines, changing machines, etc., are sorted according to their denominations on the one hand and / or sorted between originals and copies or counterfeits thereof on the other hand. Various methods of performing this classification are known. An example is described in GB2238152A, the contents of which are hereby incorporated by reference. For example, measurements representing di...

AI Technical Summary

Problems solved by technology

[0010] The problem with NCA is that the dimensionality of the nonlinear space can be very large and thus the number of principal components

Images