Efficient Banknote Recognition Based on Selection of Discriminative Regions with One-Dimensional Visible-Light Line Sensor.

Tuyen Pham,Seung Kwon,Dae Jeong,Kang Park,Sungsoo Yoon,Young Park

doi:10.3390/s16030328

Abstract

Banknote papers are automatically recognized and classified in various machines, such as vending machines, automatic teller machines (ATM), and banknote-counting machines. Previous studies on automatic classification of banknotes have been based on the optical characteristics of banknote papers. On each banknote image, there are regions more distinguishable than others in terms of banknote types, sides, and directions. However, there has been little previous research on banknote recognition that has addressed the selection of distinguishable areas. To overcome this problem, we propose a method for recognizing banknotes by selecting more discriminative regions based on similarity mapping, using images captured by a one-dimensional visible light line sensor. Experimental results with various types of banknote databases show that our proposed method outperforms previous methods.

Highlights

The accurate and reliable recognition of banknotes plays an important role in the growing popularity of payment facilities such as automatic teller machines (ATM) and currency-counting machines
Using the Fisher criterion in linear discriminant analysis (LDA), our goal is to find the optimal number of principal component analysis (PCA) dimensions for banknote feature extraction so that the following ratio is maximized: F“
We used a database consisting of 99,236 images captured from 49,618 United States dollar (USD) banknotes on both sides

Summary

Introduction

The accurate and reliable recognition of banknotes plays an important role in the growing popularity of payment facilities such as ATMs and currency-counting machines. Wu et al [1] proposed a banknote orientation recognition method that uses the average brightness of eight uniform rectangles on a banknote image as the input of the classifier using a three-layer back-propagation (BP) network. A Chinese banknote recognition method using a three-layer neural network (NN) was proposed by Zhang et al [2] This method uses linear transforms of gray images to reduce the effect of noise and uses the edge characteristics of the transformed image as the input vectors to the NN classifier. This method was applied to Sri Lankan banknote recognition in [3].

Objectives

Methods

Results

Conclusion