Klasifikasi Citra Mata Uang Rupiah Logam Untuk Deteksi Nominal Menggunakan Algoritma Convolutoinal Neural Network

Dicky, Revan Pangestu (2022) Klasifikasi Citra Mata Uang Rupiah Logam Untuk Deteksi Nominal Menggunakan Algoritma Convolutoinal Neural Network. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text cover.pdf Download (525kB)
	Text Abstract.pdf Download (5kB)
	Text Abstrak.pdf Download (6kB)
	Text BAB 1.pdf Download (89kB)
	Text BAB 2.pdf Download (420kB)
	Text BAB 3.pdf Download (122kB)
	Text BAB 4.pdf Restricted to Registered users only Download (344kB) | Request a copy
	Text BAB 5.pdf Download (6kB)
	Text Daftar Pustaka.pdf Download (157kB)
	Text lampiran.pdf Restricted to Registered users only Download (45kB) | Request a copy

Abstract

Money is a transaction tool that can be in the form of pieces of metal or pieces of paper and have different values. Money is an item that is needed by all human beings as a means of transactions in buying and selling goods or services. This study was made to classify the image of coins which can later be made an application for the detection of coins. The method used in this study is Deep Learning with the algorithm used Convolutional Neural Network (CNN) and the model used is very common, namely the sequential model. CNN is a type of neural network that is commonly used in image data processing. This study uses the CNN algorithm because this algorithm is good for classifying images and objects. The images used in this research are a total of 576 image datasets which consist of 4 classes, namely, coins of Rp. 100, Rp. 200, Rp. 500, and Rp. 1000. This model uses a learning rate parameter of 0.0001, kernel 3x3, steps per epoch 80, and there are 3 iterations used, namely 20 epochs, 25 epochs, and 30 epochs. The number of images tested were 58 images, in the test of the three iterations the greatest accuracy obtained was in the 30th epoch which got the highest accuracy of 89.5%. Keywords: Classification, Coins, CNN, Deep Learning, Epoch

Actions (login required)

View Item