An Automated Recognition of Fake or Destroyed Indian Currency Notes in Machine Vision
ABSTRACT
In
this paper, we present a microcontroller based system to automatically identify
the serial numbers of the Indian currency. The system utilizes several image
processing techniques and algorithms. The system is accurate, fast and reliable
and low cost in comparison with currently available systems.
Almost every country in the world face
the problem of counterfeit currency notes, but in India the problem is acute as
the country is hit hard by this evil practice. Fake notes in India in
denominations of Rs.100, 500 and 1000 are being flooded into the system. In
order to deal with such type of problems, an automated recognition of currency
notes in introduced by with the help of feature extraction, classification
based in SVM, Neural Nets, and heuristic approach. This technique is also
subjected with the computer vision where all processing with the image is done
by machine. The machine is fitted with a CDD camera which will scan the image
of the currency note considering the dimensions of the banknote and software
will process the image segments with the help of SVM and character recognition
methods. ANN is also introduced in this paper to train the data and classify
the segments using its datasets. To implement this design we are dealing with MATLAB
Tool.
Download PDF
Download PDF
