Banknote Recognition in Real Time Using ANN

Ren, Yueqiu
Yan, Wei Qi
Item type
Degree name
Master of Computer and Information Sciences
Journal Title
Journal ISSN
Volume Title
Auckland University of Technology

Financial institutions have adopted various automated banking systems using currency recognition as their main activity, which makes automated currency recognition of significant interest. However, after the review of the literature related to banknote recognition, it turns out that there has not been found any methods implemented or proposed for the recognition of the newly released banknotes. This thesis investigates various methods for achieving banknote real-time recognition using digital image processing. The new Series 7 New Zealand banknotes are considered as an example for intelligent banknote recognition in real time. Several experiments have been conducted in this study and two groups of training datasets are generated for comparison. One group is composed of banknote images produced by using scanners, and the other group is made up of banknote images captured by webcam. Various combinations of extracted features and classifiers have been analysed. The corresponding results are compared and the performance of each combined method is evaluated. Eventually, the PCA-based composite feature together with the BPNN is the combined method proposed in this thesis. The proposed method has demonstrated excellent performance and comparatively less time-consumption that makes it suitable for real-time applications. To the best of our knowledge, the composite feature containing both colour and texture elements, presented in this thesis has appeared in the field of banknote recognition for the first time. Our contribution is that this research project fills the vacancy of the real-time recognition of the newly released banknotes; and the proposed method paves the way for the future development of multi-currency real-time recognition.

Series 7 New Zealand paper currency , Real-time banknote recognition , Uniform LBP , Back-propagation neural network , F-measure
Publisher's version
Rights statement