Abstract:
Counterfeiting is one of the critical problems affecting cash transactions. Counterfeit banknotes
are becoming serious threats hampering the smooth transactions in Ethiopia. Hence, the
availability of such counterfeit notes in the market needs the automation of money identification
system. The importance of automatic methods for currency recognition has been increasing in the
time being because of circulation of fake notes increased in today’s economy. This recognition
system contains basic image processing techniques such as image acquisition, image
preprocesses, feature extraction and identification using Hough Transform. Counterfeit currency
identification was carried out on wide stripe and check for blind security features, which involved
employing the appropriate image preprocessing algorithms to enhance the input image applying
HT technique to extract security features and evaluating the performance of the Hough Transform
algorithm applying on both genuine and counterfeit Ethiopian banknotes. Training, validation and
testing was carried on 120 genuine and counterfeit currency notes. The results showed that using
the proposed algorithm the identification rate for new, old and very old notes is 100%, 84% and
80% respectively.
