Klasifikasi Citra Uang Kertas Rupiah Asli Dan Uang Kertas Mainan Untuk Membantu Tunanetra Dengan Metode Convolutional Neural Network

Puspa, Rahmawati (2022) Klasifikasi Citra Uang Kertas Rupiah Asli Dan Uang Kertas Mainan Untuk Membantu Tunanetra Dengan Metode Convolutional Neural Network. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text Cover.pdf Download (321kB)
	Text Abstract.pdf Download (7kB)
	Text Abstrak.pdf Download (7kB)
	Text BAB I.pdf Download (24kB)
	Text BAB II.pdf Download (652kB)
	Text BAB III.pdf Download (581kB)
	Text BAB IV.pdf Restricted to Registered users only Download (2MB) | Request a copy
	Text BAB V.pdf Download (9kB)
	Text Daftar Pustaka.pdf Download (86kB)
	Text Lampiran.pdf Restricted to Registered users only Download (38kB) | Request a copy

Abstract

Blind persons are the most common victims of counterfeit money users because they cannot see the counterfeit money and can only feel it through a blind code. These issues occur frequently and result in significant loss for blind people. One solution that can help to mitigate this issue is to develop a system that can detect the authenticity of money. In this study, we used the Deep Learning method with the Convolutional Neural Network. Jupyter Notebook is used to write Python programming code scripts while building the model. The dataset used in this study includes images of the original Rp 50,000 and Rp 100,000 banknotes, as well as 800 images of the Rp 50,000 and Rp 100,000 toy banknotes. The image is split into three folders: data train, validation, and test. Epochs of 10, 20, 30, 40, 50, 60, 80 and 100 were used in the test model. According to the test results, the accuracy value on the train and validation data was up to 100 percent in all test models. The Predicted class test results with the actual class on 80 test images produced correct predictions, indicating that this system was successful. Keywords : Currency, Fake Currency, Blind, Deep Learning, CNN, Epoch, Accuracy

Actions (login required)

View Item