Implementasi Deep Learning Berbasis Tensorflow Untuk Pengenalan Sidik Jari

Nurfita, Royani Darma and -, Gunawan Ariyanto,ST. M. Compt Sc., Ph.D. (2018) Implementasi Deep Learning Berbasis Tensorflow Untuk Pengenalan Sidik Jari. Skripsi thesis, Universitas Muhammadiyah Surakarta.

[img] PDF (Naskah Publikasi)

Download (2MB)
[img] PDF (Surat Pernyataan)
Surat Pernyataan.pdf
Restricted to Repository staff only

Download (366kB) | Request a copy


The fingerprint recognition system is widely used in biometrics for various purposes in recent years. Fingerprint recognition is used because it has a complex pattern that can recognize a person and is the identity of every human being. Fingerprints are also widely used as verification and identification. Problems encountered in this research is the difficult to classify objects one of them on fingerprints. In this study the authors use deep learning using the method of Convolutional Neural Network (CNN) to overcome these problems. CNN is used to perform machine learning process on computer. Stages on CNN are data input, preprocessing, training process. The implementation of CNN used in this research is tensorflow library by using python programming language. The dataset used originated from a fingerprint verification competition website in 2004 using optical sensor type “V300” by CrossMatch and in it there were 80 fingerprint images. The training process uses 24x24 pixel data and performs the test by comparing the number of epoch and learning rate so it is known that if the greater the number of epoch and smaller the learning rate the better the accuracy of the training obtained. In this research, the accuracy level of training is 100%.

Item Type: Karya ilmiah (Skripsi)
Uncontrolled Keywords: Pengenalan sidik jari, Deep Learning, Convolutional Neural Network, Tensorflow, Python
Subjects: T Technology > T Technology (General)
Divisions: Fakultas Ilmu Komunikasi dan Informatika > Teknik Informatika
Date Deposited: 07 Jul 2018 03:59
Last Modified: 07 Jul 2018 03:59

Actions (login required)

View Item View Item