A Hierarchical Classification Method for US Bank-Notes

Abstract

This paper describes a method for the classification of bank notes. The algorithm has three stages, and classifies bank notes with v ery low error rates and at high speeds.To achiev e the v ery low error rates, the result of classification is check ed in the final stage by using different features to those used in the first two. High-speed processing is mainly achiev ed by the hierarchical structure, which leads to low computational costs. I n ev aluation on 32, 850 samples of US bank notes, with the same number used for training, the algorithm classified all samples precisely with no error sample.

A High Performance Banknote Recognition System Based on a One-Dimensional Visible Light Line Sensor

“…The images in the database consisted of the four directions for the 16 types of banknotes ($1, $2, $5, recent $5, most recent $5, $10, recent $10, most recent $10, $20, recent $20, most recent $20, $50, recent $50, most recent $50, $100, recent $100). As shown in Table 2 , the number of images used in our study is similar to that of [ 10 ], but much larger than in most previous studies [ 11 , 12 , 13 , 14 , 15 , 16 , 17 ]. In addition, our study uses more classes than any previous study [ 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 ].…”

“…In the next experiment, we measured the processing time for our method using a desktop computer (a 3.5 GHz CPU with 8 GB RAM). Most previous research on banknote recognition does not measure the processing time, focusing exclusively on the recognition accuracy instead [ 9 , 10 , 12 , 13 , 14 , 15 , 16 , 17 ]. Therefore, it is difficult to compare the processing time of our method with others.…”

“…Previous studies can be divided into those that address the recognition of a banknote’s orientation [ 9 ] and those concerned with banknote recognition [ 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 ]. Wu et al .…”

“…proposed a hierarchical classification method for United States Dollar (USD) banknotes. Their method consists of three stages, using a generalized learning-vector quantization (GLVQ) algorithm to achieve high-speed processing with a high degree of accuracy [ 10 ]. The performance of their method was such that 99% of the banknotes were correctly recognized.…”

Classification of Mexican Paper Currency Denomination by Extracting Their Discriminative Colors

“…Related works on paper currency recognition have focused on the classification of Euros [4], [5] and US Dollars [6], [7]. Other related works address the recognition of the paper currency of specific countries [8]- [12], in these papers the banknotes of each country are modeled by extracting their very particular features; so, the recognition methods are customized depending on the nationality of the paper currency.…”