Abstract
Security systems that use passwords or identity cards can be hacked and misused. One of alternative security system is to use biometric identification. The biometric system that is popularly used is fingerprints, because the system is safe and comfortable. Fingerprints have a distinctive pattern for each individual and this makes fingerprints relatively difficult to fake, so the system is safe. Comfortable because the verification process is easily done. The problem that often occurs on the system of fingerprint scanner is found an error and the user has difficulty when accessing. To handle with these problems has developed an artificial intelligence system. One of arificial intelligence in pattern identification is artificial neural networks (ANN). From some of the results of previous research showed that the ANN method is reliable in pattern identification. Based on these facts, the method used in this research is the perceptron ANN method with values learning rate varying. In the research the program conducted by testing 20 samples showed that the performance of the perceptron ANN method is relatively good method in fingerprint image recognition. This can be indicated from the value of accuracy (0.95), precision (0.83), TP rate (1), and FP rate (0.07)). In addition, the location of the point coordinate (FP rate; TP rate) is (0.07; 1) in ROC graphs is located on the upper left (perfect classifier region).
Highlights
Security systems that use passwords or identity cards can be hacked and misused
comfortable. Fingerprints have a distinctive pattern for each individual
The problem that often occurs on the system of fingerprint scanner is found an error
Summary
1,2Program Studi Fisika, Fakultas Matematika dan Ilmu Pengetahuan Alam, Universitas Udayana, Kampus Bukit, Jimbaran, Badung, Bali, Indonesia 80361. Bedasarkan pada fakta tersebut, dalam penelitian ini digunakan metode JST perceptron dengan nilai learning rate bervariasi. Dari hasil penelitian yang dilakukan dengan menguji 20 sampel menunjukkan bahwa kinerja metode JST perceptron merupakan metode yang relatif baik dalam pengenalan citra sidik jari. Untuk menangani permasalahan tersebut telah berkembang sistem kecerdasan buatan (artifical intelegent), dimana salah satu bidang kecerdasan buatan dalam identifikasi pola adalah jaringan saraf tiruan (JST) [1]. Pengenalan sidik jari menggunakan metode tersebut menghasilkan nilai error yang kecil, yaitu kurang dari 0,0005 [4]. Oleh karena itu dalam penelitian ini menggunakan metode JST perceptron untuk pengenalan citra sidik jari. Ukuran yang dipakai untuk menilai kinerja dari metode JST perceptron yaitu: keakuratan prediksi, percision, recall, false positive rate, dan ROC graphs. 3.2 Alur Penelitian Langkah yang dilakukan dalam pengambilan dan pengolahan sampel ditujukkan pada Gambar 3
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
