ABSTRAKSI: Iris is the annular region of the eye bounded by the pupil and the sclera (the white part of the eye). Visual texture of the iris is formed during fetal development and stabilize itself during the first two years of the life of the fetus. Complex iris texture brings a very unique and useful informaItion for a personal recognition. The speed and accuracy of the Iris-based recognition system is very promising and it is possible to use on a large scale identification systems. Each iris is unique and like fingerprints, iris texture is different from identical twins. The texture of the iris is very difficult to surgically destroyed.
Dalam tugas akhir ini dibangun sistem yang mampu mengenali iris dalam bentuk citra digital. Citra digital yang di dapat dari kamera digital dilakukan deteksi menggunakan metode Local Binary Pattern (LBP), setelah itu diklasifikasikan dengan menggunakan jaringan syaraf tiruan Self Organizing Maps (SOM).
Pada tugas akhir ini telah dibuat suatu aplikasi yang dapat mengidentifikasi seseorang melalui pola iris matanya. Digunakan metode local binary pattern sebagai ekstraksi ciri dan JST-SOM untuk pengenalan pola dari citra iris mata. Citra mata yang diproses adalah data dari Chinese Academy of Sciences –Institute of Automation(CASIA). Dengan menggunakan parameter-parameter yang optimum, sistem ini mampu menghasilkan performansi yang cukup baik karena mampu menghasilkan nilai akurasi maksimum 95% dan waktu komputasi rata-rata selama 0.01379 detik untuk setiap citra iris.
Kata Kunci : Iris, JST, SOM, Local Binary Pattern, CASIAABSTRACT: Iris is the annular region of the eye bounded by the pupil and the sclera (the white part of the eye). Visual texture of the iris is formed during fetal development and stabilize itself during the first two years of the life of the fetus. Complex iris texture brings a very unique and useful informaItion for a personal recognition. The speed and accuracy of the Iris-based recognition system is very promising and it is possible to use on a large scale identification systems. Each iris is unique and like fingerprints, iris texture is different from identical twins. The texture of the iris is very difficult to surgically destroyed.
In this final project built a system capable of recognizing the iris in the form of a digital image. Digital image on a digital camera can do detection using Local Binary Pattern (LBP), and then classified using a neural network Self Organizing Maps (SOM).
In this final project has created an application that can identify a person through iris patterns. Used as a method of local binary pattern feature extraction and ANN-SOM for pattern recognition of iris images. Eye image is processed data from the Chinese Academy of Sciences-Institute of Automation (CASIA). By using the optimum parameters, the system is able to produce a pretty good performance because it can produce maximum accuracy value of 95% and the average computation time for 0.01379 seconds on each image slice.
Keyword: Iris, ANN,SOM, Local Binary Pattern, CASIA