International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 Volume: 08 Issue: 05 | May 2021 www.irjet.net p-ISSN: 2395-0072 © 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 880 Currency Authentication Using Image Processing Prof. Manisha Singh 1 , Shubham Ambike 2 , Vishal Bhat 3 , Prince Sunny 4 , Nadeem Shaikh 5 , Deepak Vaishnav 6 1 (Asst. Professor, Dept. of Computer Engineering, DPCOE, Pune, Maharashtra) 2, 3, 4, 5, 6 (Student, Dept. of Computer Engineering, DPCOE, Pune, Maharashtra) ---------------------------------------------------------------------***---------------------------------------------------------------------- Abstract - The characteristics of paper notes vary from country to country. Currency authenticity or say identification is one among the important applications of pattern recognition. We have proposed a system for the automation of currency recognition using image processing techniques. The proposed system can be used for recognizing as well as authenticating given Indian banknotes. Only paper currency will be considered. This method works by identifying certain predefined areas of interest, and then extracting the denomination value using various characteristics such as color and text on the note. Our system identifies currency quickly and accurately. Initially, our system will be taking the frontside and backside of the currency(your note) as an input and then crop it into specific predefined areas of interest. Then each image is divided into three channels. Filtering is applied to each channel, the red, green, and blue channels are recombined to get back the RGB image. Different features such as HSV are extracted from the RGB image. The proposed model is based on a feature extraction and k- Nearest Neighbor (k-NN) classifier for recognizing test banknote. The recognition system indicates that the proposed approach is one among the foremost effective strategies for identifying currency patterns to read its face value and determine its authenticity for Indian currency notes. Key Words: Indian Currency Recognition, Feature Extraction, HSV, k-Nearest Neighbour. 1.INTRODUCTION The identification of currency depends on the characteristics of currency notes of a particular country. Due to use for a long time, currency notes may be contaminated by noises. To Identify whether the currency is authentic or not there are many features. Although it may not be practically possible to accurately identify a counterfeit in a paper currency which can only be identified by an intelligent machine. Modern automation systems in the real world requires a system that will recognize currency. It has various potential applications that includes banknote counting machines, money exchange machines, assisting blind persons, electronic banking, currency monitoring systems etc. The recognition of currency is a very important need for visually impaired people. They are not being able to differentiate between currencies correctly, so it is very easy for them to be cheated by the others. Therefore, there is an urgent need to design a system that will recognize the currency authenticity and its value. 2. CURRENCY DETECTION 2.1 Identifying distinguishable region Once the pre-processing steps have been done, we can identify which regions of the note are relatively distinguishable. This is done based on certain predefined areas. All the currencies have certain regions with fixed coordinates which we crop out for classification of notes. We have chosen 5 features from each currency note that are highlighted below within the images. Grouping is completed by checking out the values for pictorial features like hue and saturation for every feature then those values are compared with the original pictures of currency. The values chosen to classify the notes are found experimentally. Fig. 2.A: Frontside of Rs 2000 banknote Fig. 2.B: Backside of Rs 2000 banknote