Abstract:
The principle reason for the creation of this system is the need for the
recognition ofthe local currency notes ofour country. Forthe desired purpose we have
collected all the currency start from the note ofRS 10 to RS 5000. The pictures will
be compared with the pictures stored in dataset and ifthe descriptors and key points
for the scanned pictures matches with the pictures in our dataset, the details ofthat
specific currency will be sent to the system. Some outputs are going to be available
whichTl show that there is good accuracy ofrecognition ofour local currency. Starting
from the noise removal techniques to make the picture look more bright and clear to
adjusting all the images to some exact size, all the possible ways were implemented to
maintain quality standard. Furthermore if we talk about the methods utilized in this
project for the feature extraction techniques or the classification, then there were a lot
ofmethods which were checked, implemented and verified and then one ofthe method
got finalized. The finalized method for feature extraction is MaxPooling2D which is
a library ofpython which can extract features from the vision of any image, it is one
of the best feature extraction technique. The technique used for classification is
Artificial Neural Networking (ANN). Itscans the picture into matrix forms, matrix are
basically the small parts ofimage, then it further gives the image to the neural network
to perform the further classification over here it has done the matching process. The
scanned image gets compared with the images present in the project’s dataset. It gives
the maximum accuracy in the results, the maximum accuracy got receivedjust by using
these mentioned techniques. As during the pandemic it’s better to use this system
instead of making human interaction as the virus spreads with the human touch and
it’s been noticed that currency is one ofthe major cause ofspreading the virus. This
system can be placed at banks and people can pay their bills at the banks using this
cashier less system. Otherthan that it also reduces human efforts asthe machine would
be able to perform some ofthe cashier’s work which can help the cashier manage their
time and work more efficiently.
