Fake currency detection using image processing

ABSTRACT:-

The Advancement in color printing technology has increased the making of fake currency notes and making of duplicate notes on a large scale. In the past few years printing was done in the printing house, but now anyone has a printer at home and can print the fake notes with maximum accuracy. This results in the issue of fake currency instead of real currency found in our market. The issues of fake currency are evident in daily reports of fake currency frauds. The enhancement in technology like computers, scanners and copies has made it much easier to create fake currency and there is no software to detect the fake or real currency.

India is surrounded by many problems like corruption and black money and fake currency are also major problems. This problem leads to creation of fake currency in less time and more efficiently. So to overcome this problem we design a system that can classify the currency into real or fake currency. This system describes the various aspects of Indian currency. There are machines available in banks and other markets to check financial authenticity. But the common man does not check each note whether fake or real and therefore the need for software to obtain fake money arises, which can be used by ordinary people. This proposed system uses image processing to determine if the currency is fake or real. The program is designed entirely using Python language. Contains steps such as grayscale conversion, edge detection, splitting, etc. which are made using appropriate methods. CNN is used to extract note features. The proposed system has advantages such as simplicity and high performance. The result will predict whether the currency note is fake or not.

OBJECTIVES OF THE PROJECT:-

The objectives can be described in two stages

Designing an algorithm for feature localization.
Designing a feature extraction and recognition.
Designing an optical character recognition for value detection.

EXISTING SYSTEM:-

From the observation of the papers we will say that there are positive stages which can be very essential in the current gadget architecture. Firstly we’ve the step known as image acquisition manner we have to take enter because the photograph most effective via the scanner and on this there may be little need of any virtual digital camera to size of the photograph withinside the actual time gadget.In this existing architecture, most effective the the front a part of the note is think about and now no longer the rear part. After that we’ve a subsequent step referred to as a pre-processing method. In this there are essentially three to four sub ranges concerned like pre-processing, grayscale conversion, edge detection and segmentation.

PROPOSED SYSTEM:-

The proposed system consists of the benefits of the existing system and removes its disadvantages. This system focuses on the improvement and implementation of the fake currency detection application. The scope of the project is to provide techniques and methods that appear suitable while you access the image of the currency you want.

In the proposed system, We work on the image of currency notes captured by the digital camera. The working of our proposed system is as follows : Firstly we capture the image of the currency note by digital camera or scanner under the ultraviolet light. Then the RGB image is converted into the grayscale image. Then the whole grayscale image is passed through the process called Edge detection. Edge detection is a process in which identification of points in a digital image with discontinuities, simply to say, sharp changes in the image brightness. This image is further processed and edges of gray scale images are detected. Image segmentation is the process of dividing an image into multiple parts by cropping it. Then the currency note features are cropped and segmented and these features are extracted. The intensity of each extracted feature is calculated and if the intensity is greater than the average value the currency note is said to be real otherwise it is said to be fake.

The characteristics used to authenticate currency notes are Security Thread, Serial Number, Latent image, Watermark, Identification Mark. The results are displayed on web UI which shows the extracted features of currency notes. The real currency note extracted features displays at least 70 percent intensity, it is seen that the 500-2000 note displays intensity less than 75 percent for some features hence it is considered as fake note.

MODULES:-

User Register and Login: User can register to check whether that currency is fake or not.
Currency Feature localization and analysis: With the help of CNN algorithm and Open-CV currency will be captured by camera and feature extraction of image will be done which will result in obtaining output such as fake or not.
CNN Model generation: Feeding currency image data to train CNN models and extracting the features.

ADVANTAGES :-

The application will prove very beneficial to detect fake currency.
The application is user friendly and easily accessible.
It will save time, reduce the effort of the user.
Provide cheaper and accurate systems to the user which can be easily accessible and give accurate recognition of currency notes.
It can be applicable to every economy level.

HARDWARE AND SOFTWARE REQUIREMENTS

HARDWARE:

Processor: Intel Core i3 or more.
RAM: 4GB or more.
Hard disk: 250 GB or more.

SOFTWARE:

Operating System : Windows 10, 7, 8.
Python
Anaconda
Spyder, Jupyter notebook, Flask.
MYSQL