KLASIFIKASI POLA SIDIK JARI MENGGUNAKAN JARINGAN SYARAF TIRUAN BACKPROPAGATION; FINGERPRINT PATTERN CLASSIFICATION USING ARTIFICIAL BACKPROPAGATION NEURAL NETWORK

SULISTIYASNI, Edi Winarko

2013 | Disertasi | PROGRAM STUDI S2 ILMU KOMPUTER

This reseach describes the fingerprint classification. Proposed to classify human based on three classes such as: whorl, arch, and loops. The proposed system consist of five steps preprocessing, segmentation, feature extraction and classification. In preprocessing there are some of steps such as grayscale, median filter, auto contras, and histogram. Segmentation used otsu thresolding method and features extraction used gray level coocurence matrix (GLCM), in wich the features are correlation, contrast, energy, homogeneity, and entropy. These classification use backpropagation neural network. The result shown that system can classify fingerprint with accuracy 87,5%.

Kata Kunci : GLCM; backpropagation neural network