Currency Recognition System Using Image Processing

1y ago
6 Views
2 Downloads
1.77 MB
7 Pages
Last View : 23d ago
Last Download : 2m ago
Upload by : Camden Erdman
Transcription

International Journal of Scientific & Engineering Research Volume 11, Issue 4, April-2020ISSN 2229-55181595Currency Recognition System Using ImageProcessingDr. Ritu Sindhu, Pranshu VermaDepartment of Computer Science and EngineeringGalgotias UniversityGreater Noida, U. P., 00@gmail.comAbstract--In this paper, we proposed an automaticcurrency recognition system using digital imageprocessing methodology. The following projectmainly focuses on the recognition of currency by itsimage or photograph. It will help users to recognizedetails about currency like Currency Value,Currency Name, the value in INR, EURO and USDollar. It works using the main characteristics ofcurrency notes such as size colour or printed text onit and also depends on differ in currency notes withinthe same country. We have considered INDIANRupee and US Dollar, the most used currencies inour domain with their denominations. This systemworks accurately and also able to quickly identify thecurrency notes.We used INR and USD as an example to illustrate thetechnique. This system is based on our knowledge aboutcomputer science technologies like Digital ImageProcessing, python and also a small step to implement ina system that is most important for industrialdevelopment.We had considered INDIAN Rupee and USDollar for this project. Project can expand for morecurrencies inclusion according to use.The block diagram as in Fig. 1 shows the usedmethods in the system to obtain the actual output sameas the expected output. First of all input a currency noteimage, which needs to be checked. Then systemperforms basic image processing techniques on inputimage and refined to convert it into a usable input formatching with present image in dataset. The systemextracts the information from image based on featuressuch as colour and texture. With the help of thesefeatures the system determines currency name anddenomination of the currency note. After this theexchange rate of currency will be extracted from internetwith the help of online exchange rate api url. At the endoutput displayed on the screen.IJSERKeywords—Image Processing, BGR to Greyscale,Binary Threshold, Currency Name, CurrencyDenomination (Value), Trained Dataset, CurrentExchange Rate.I. INTRODUCTIONAccording to the UN charter there are around195 countries around the globe. In which 193 countriesare members of the UN and two are observing states.According to The U.N., worldwide there are 180currencies. All these currencies are different incharacteristics such as size, colour and texture. In the eraof rapidly growing levels of trade between countries andalso tourism all over the world, it becomes necessary torecognize each currency note correctly. Now a dayspeople travel to different countries, they use their nativecountry currency in paying bills or buying stuffs andbecause most of the local people are not familiar withthe currency other than their own country currency andalso not familiar with the exchange rate of that currencyin their own currency, it becomes necessary to developan automated system that helps in recognition noteseasily, faster with efficiency.The proposed system is based on imageprocessing and makes the process robust and automatic.A. RELEVANCEThe proposed system is built with the motive tohelp common people for currency recognition at anytimeand anywhere. When people see a new currency thatdoesn’t belong to their native country then they facemany issues in recognition that currency’s value, origincountry and also face difficulties in finding the exchangerate. It is need to develop a system which will helppeople to identify currency value, currency name andexchange rate only by providing currency image.In this paper we proposed an automation systemfor easily recognition of currency. The system usesdigital image processing techniques to extractinformation from inserted image and match with thetemplate images.IJSER 2020http://www.ijser.org

International Journal of Scientific & Engineering Research Volume 11, Issue 4, April-2020ISSN 2229-5518To implement image processing techniques weused python as programming language because it ismuch easier to implement digital image processing byusing python because of its pre-defined packages and1596frameworks. NumPy and OpenCv are the twoframeworks which are used to perform image processingfunctionalities and Tkinter is used to design applet of theapplication.Fig. 1 Block diagram of the Proposed SystemII. LITERATURE SURVEY[1] In 2010, ‘K. K. Debnath et al’ proposed a systemfor currency recognition using Ensemble NeuralNetwork (ENN). In this model, negative correlationlearning is used to train the Individual Neural Networks(NNs) in an ENN. They used different types of notessuch as noisy and old notes because the machine doesn’teasily recognize such notes. That’s why ENN is used todevelop a system to identify them easily and correctly.For testing, they used different dominations notes, whichare of 2, 5, 10, 20, 50, 100 and 500 TAKA. The systemfirstly converts the note image into greyscale and thenthe image is compressed. After compression, thecompressed image is processed to the system as an inputfor recognition. Due to ENN, the system can easilyidentify the currency with noise as well as old currencynotes. The system has fewer chances of misclassificationbecause of independent training.The trained data set is created using the backpropagationalgorithm. Axis symmetric mask is used in the preprocessing stage for correct recognition of currencynotes. They used eight TAKA notes for testing thesystem and the result was successfully recognition of allnotes.IJSER[2] In 2010, ‘J. Guo et al’ proposed a model thatuses the Local Binary Pattern (LBP) algorithm for papercurrency recognition. It extracts currency features withgood quality for efficient recognition of currency notes.LBP algorithm is used for the extraction ofcharacteristics of a currency. LBP algorithm is based onthe LBP method and the LBP method has the advantageof simplicity and high speed.[3] In 2011, ‘Jahangir and Raja’ proposed a systemto recognize Bangladeshi currency with the help of theneural network recognition method. This method hadimplemented on cost-effective hardware so that can beused in different places. The system takes a currencynote image as input and notes are scanned using sensors.[4] In 2013, ‘Qing and Xun’ proposed a modelwhich was based on Artificial Neural Networks andGene Algorithm. ANN and Gene algorithm are problemsolving techniques. Gene Algorithm had used because ofslow convergence and indeterminate initial weights forbackpropagation neural networks. The Gene Algorithm:Back Propagation takes short training time with fasterrecognition speed and provides the appropriate result ofconnection weights and network connection.[5] In 2000, ‘Y. Mitsukura et al’ proposed a modelthat depends on Artificial Intelligence methods andadaptive systems, combining Neural Network methodswith the help of the Simulated Annealing Method. Theproposed system was low cost with a recognition rateresult of 99.68%. The result was achieved by developingsmall size neural network for coin recognition with thehelp of a genetic algorithm and simulated annealing. III. OBJECTIVE of the SYSTEMThe main objective of this system is to recognizecurrency using image processing technique.The system performs image processing on insertedimage and match with trained data set.To save user time it needs device should beconnected to internet so that it can extract currentIJSER 2020http://www.ijser.org

International Journal of Scientific & Engineering Research Volume 11, Issue 4, April-2020ISSN 2229-5518 exchange rate of that currency into other mainlyused currencies.For using this no need for high specs hardware andeasy to use.IV. PROPOSED MODELFlowchart for proposed system Fig. 2:1. Start2. Give input as image (JPG/PNG/GIF format).3. Conversion of image from BGR to Greyscale.4. The Binary Threshold value will be assigned withthe pixels of greyscale image.5. Match the greyscale image with template image.6. If image matched with the trained data set then showthe defined output. And if not matched with anyimage in data set then return error with message.7. Stop the processA. METHODOLOGYFor Image processing and Template matching:1. img cv2.imread(filename) /*for taking inputimage and read that image*/2. gray img cv2.cvtColor(img,cv2.COLOR BGR2GRAY) /*conversion ofinput image from BGR to Greyscale.*/3. temp "0" /*initializing a variable temp with 0value*/3.1. entries os.listdir("trained data/") /*obtaininglist of all files/directories from trained dataset.*/4. for entry in entries:4.1. template cv2.imread("trained data/" entry,cv2.THRESH BINARY) /*Assigning binarythreshold value to the entered greyscale imagepixels, 1 for white pixels and 0 for blackpixels.*/4.2. w, h template.shape[::-1] /* w width andh height and template.shape used to find andsearch location of a template image in greyscaleimage.*/4.3. result cv2.matchTemplate(gray img, template,cv2.TM CCOEFF NORMED) /*compare atemplate with the overlapped image regions.*/4.4. loc np.where(result 0.4) /* returns indicesof image where matching satisfies.*/4.5. for pt in zip(*loc[::-1]):cv2.rectangle(img, pt, (pt[0] w, pt[1] h), (0,255, 0), 0)4.6. temp entrybreak4.7. imageid temp.replace(".jpg","")print(" Recognized with Image Id : " imageid)5. if imageid is None: /* if image doesn’t match withany image in trained dataset.*/5.1. window.windows.error("SorryImageNotRecognised") /* gives error with message inapplet window.*/6. else:6.1. imageid int(imageid) – 1 /* and when imagematch with template image.*/6.2. if currencydata[int(imageid)]['currency'] "Indian Rupee": /* if inserted image matchedwith Indian Currency template image.*/url https://api.exchangeratesapi.io/latest?base INR/* url for getting live exchange rate in USD andEuro.*/6.2.1. response requests.get(url)/*requesting exchange rate from urlserver.*/IJSERFig. 2 Flowchart of the system1597IJSER 2020http://www.ijser.org

International Journal of Scientific & Engineering Research Volume 11, Issue 4, April-2020ISSN .3. else:6.3.1.6.3.2.data response.text /* reading responsedata text from server.*/parsed json.loads(data) /* Loadingparsed for using json data into python.*/date parsed["date"] /*parsing ofdate.*/eur rate parsed["rates"]["EUR"] /*parsing of Euro exchange rate w r tINR.*/usd rate parsed["rates"]["USD"] /*parsing of USD exchange rate w r geid)]['value'])/*for printing currency value on int(imageid)]['currency'])/*for printing currency name on outputscreen.*/print("Currency value In EUR:" ,eur rate*currencydata[int(imageid)]['value']) /* for printing exchange rate inEURO on output screen.*/print("Currency value In USD:" ,usd rate*currencydata[int(imageid)]['value'] ) /* for printing exchange rate inUS Dollar on output screen.*/currvalue currencydata[int(imageid)]['value']/*obtaining currency value from traineddataset.*/currname currencydata[int(imageid)]['currency'] /*obtaining currency namefrom trained dataset.*/curreur eur rate*currencydata[int(imageid)]['value'] /*For getting currencyexchange rate with respect to euro. */currusd usd rate*currencydata[int(imageid)]['value']/*printing current exchange rate withrespect to USD.*/window.windows.getind(currvalue,currname, curreur, currusd,self) /* fordisplaying output on 6.3.8.6.3.9.6.3.10.1598response requests.get(url)/*requesting exchange rate from urlserver.*/data response.text /* reading responsedata text from server.*/parsed json.loads(data) ) /* Loadingparsed function for using json data intopython.*/eur rate parsed["rates"]["EUR"]/*parsing of Euro exchange rate w r tUSD.*/inr rate parsed["rates"]["INR"]/*parsing of USD exchange rate w r geid)]['value'])/*forprinting currency value on int(imageid)]['currency'])/*for printing currency name on outputscreen.*/print("Currency value In EUR:" ,currencydata[int(imageid)]['value']*eur rate) /* for printing exchange rate inEURO on output screen.*/print("Currency value In INR:" ,currencydata[int(imageid)]['value']*inr rate ) /* for printing exchange ratein INR on output screen.*/currvalue currencydata[int(imageid)]['value'] /* obtaining currency value fromtrained dataset.*/currname currencydata[int(imageid)]['currency'] /* obtaining currency namefrom trained dataset.*/curreur currencydata[int(imageid)]['value'] * eur rate /*printing currentexchange rate with respect to USD.*/currinr currencydata[int(imageid)]['value'] * inr rate /*printing currentexchange rate with respect to INR.*/window.windows.getusd(currvalue,currname, curreur, currinr,self) /*fordisplaying output on applicationwindow.*/IJSERurl "https://api.exchangeratesapi.io/latest?base USD" /* url for getting liveexchange rate in INR and or applet:1. Imported frameworks and modules Fig. 3:IJSER 2020http://www.ijser.org

International Journal of Scientific & Engineering Research Volume 11, Issue 4, April-2020ISSN 2229-551815994. ‘Reset’ button:Fig. 6 Reset buttonFig. 3 Frameworks and Modules2. ‘Open Image’ button:B. SYSTEM DESCRIPTIONInput Image: An image file having extensionJPG/PNG/GIF will be provided initially. The imageshould be of currency notes (INR, USD). It can takeUSD (1, 2, 5, 10, 20, 50, 100) and INR (1, 2, 5, 10, 20,100, 200, 500, 2000).Image Conversion:To convert image from BGR togreyscale we used OpenCv and numpy modules ofpython. It performs BGR to greyscale conversion usingdigital image processing. It performs the conversion andsets the binary threshold value for each pixel ingreyscale image to perform template matching. BinaryThreshold assigns values to every pixel in binary form.Value 1 is assigned to white pixels and 0 to black pixels.IJSERTrained data set: We have trained data set for allcurrency notes which we have considered in our system.We have around 2 to 4 samples of each currency note inour trained data set which we have considered.Fig. 4 Open Image button3. ‘Recognize’ button:Template Matching: It is the step where converted imagematched with the trained data set to recognize thecurrency and show its value. If data doesn’t match withany template then it will give error and show a message.Current Exchange Rate: We used an url to show thecurrent exchange rate in INR, USD/EURO. It will makecurrency recognition much easier for people withexchange denomination and also will save time. But theessential condition is that we need an active internetconnection and the device should be connected to theinternet. Exchange rate works with url server to extractcurrent exchange rate and multiply with the detectedcurrency denomination. Output will be print onapplication window.Fig. 5 Recognize buttonV. RESULTFirstly we run the application, welcome windowwill open as in Figure 7. Welcome window have 4buttons ‘Open Image’, ‘Recognize’, ‘Result’ and ‘Exit’button.Open Image: When you click on this button then it willdirect you to memory and ask you to select image ofIJSER 2020http://www.ijser.org

International Journal of Scientific & Engineering Research Volume 11, Issue 4, April-2020ISSN 2229-55181600currency. After selecting an image you willautomatically return to the window and your selectedimage will be visible in application window as in Fig. 9.Recognize: This button is used for currency recognition.After pressing this button currency recognition will startand you will get output in few seconds on theapplication window as in Fig. 12.Reset: This button is used to clear the applicationwindow.Exit: This works as simple exit button, you will comeout of the application and all processes of applicationwill be terminated.Fig. 9 Input ScreenDuring image processing, firstly it will convert insertedimage from BGR to Greyscale as in Fig. 10.IJSERFig. 7 Welcome windowInput Image: It is the currency image which you want torecognize as in Fig. 8.Fig. 10 Greyscale ImageAfter conversion from BGR to Greyscale it will performtemplate matching as in Fig. 11 for correctly recognitionof the currency with currency name and currency value(Denomination).Fig. 8 Input ImageInput Screen as in Fig. 9 is application screen whereyour select image will appear and allow you to performactions defined in system.Fig. 11 Template MatchingIJSER 2020http://www.ijser.org

International Journal of Scientific & Engineering Research Volume 11, Issue 4, April-2020ISSN 2229-5518The final output will contain Currency Value, CurrencyName, Value in Euro and Value in INR/Value in USD.And output will also show the country flag.10. https://www.googlescholar.com11. ing/12. https://www.elprocus.com13. https://www.worldometers.info/geography/14. cies-are-in-the-world.htmlFig. 12 Output ScreenREFERENCESIJSER1. Pearson Book on Digital Image Processing 4th2.3.4.5.6.7.8.9.1601Edition By Gonzalez (2018).Programming and Problem Solving with Python 1stEdition by ‘Ashok Namdev Kamthane’ and ‘AmitAshok Kamthane’ 2017Identification of Paper Currency Techniques: ASurvey by S. Sahu and T. Verma InternationalJournal of Science Technology & Engineering(2016)Recognition system for real time paper currency.IEEE 2012.Extraction of serial number on bank notes. IEEE.Digital Image Processing And Analysis: HumanAnd Computer Vision Applications With Cviptools,2nd Edition by Scott E Umbaugh, T and F India.An intelligent system for paper currency recognition- robust features Journal of Intelligent & FuzzySystems - A. Sargano, M. Sarfrazb, and N. Haq,2014.Paper Currency Recognition - Journal of AdvancedComputer Science and Technology Research ,‘Althafiri E, Sarfraz M’, ‘Alfarras M. Bahraini’.Study of Printing Identification Based on Multispectrum Imaging Analysis, Proceedings of theInternational Conference on Computer Science andSoftware Engineering, By Liu Q, Tang L. p. 229 –232; 2008.IJSER 2020http://www.ijser.org

many issues in recognition that currency's value, origin country and also face difficulties in finding the exchange rate. It is need to develop a system which will help people to identify currency value, currency name and exchange rate only by providing currency image. In this paper we proposed an automation system

Related Documents:

Transaction currency The currency in which a transaction originates. Accounting currency The primary currency in which a legal entity maintains its financial records. The accounting currency is stored in the Ledger table. Reporting currency The reporting currency of the ledger. The reporting currency is stored in the Ledger table. It is optional.

the currency of the bond's denomination as the local currency and the chosen currency of the portfolio or index as the base currency. The return of this security in the base currency on day t can be computed using the following inputs. and 1 are the market values in local currency at the close of day t and t-1 respectively.

Currency Exchange Setup. Adding Currency Code and Exchange Rates2-1. Editing Currency Exchange Rate2-2. Removing Currency Exchange Rate2-2. Commission Charge Code and Account2-2. Setting Up Commission Charge Code2-4. Setting up System Account2-4. Performing Currency Exchange Transaction. Buy Sell Currency Exchange3-1. Exchange Charge to Account3-2

Currency Harvest was 15.22%. The Global currency harvest returned 12.07% and the G10 Currency harvest returned 8.02%. The Sharpe ratios for the Balanced Currency Harvest, Global Currency Harvest and G10 Currency Harvest over the period were 1.60, 1.23 and 1.07 respectively. (The returns are in USD terms

a sterling/Euro currency account held with us.) Each payment up to the currency equivalent of 100 Free Each payment over the currency equivalent of 100 6 Sterling and foreign currency payments payable to an account held with us in a different currency from the payment Each payment up to 100 or the currency equivalent of 100 Free

from an underlying asset which in foreign currency options is the exchange rate. There are two types of foreign currency options, a call currency option and a put currency option. A call option on a particular currency gives the holder the right but not an obligation to buy that currency at a predetermined exchange rate at a particular date and .

1. Vipin Kumar Jain, Dr.RituVijay,"Indian currency Denomination Identification Using Image Processing Technique" IJESRT, Vol.2, Issue 1, pp.125-128, January 2013. 2. An automatic recognition of fake Indian currency note using MATLAB (IJESIT). 3. Reviews of various image processing techniques for currency note authentication (IJCERT). 4. H.

9 MATHEMATICS - Week 1 Lesson 2: Properties of Operations Learning Objectives: Students will be able to simplify computations with integers, fractions and decimals by using the associative and commutative properties of addition and multiplication, and