Open AccessIris feature extraction using Gray Level Co-occurrence Matrix and Gabor Kernel filter its impact on iris Huffman compression image
Author(s) -
Rosa Andrie Asmara,
R D R Yusron,
Faisal Rahutomo,
Rudy Ariyanto,
Deddy Kusbianto Purwoko Aji,
Priska Choirina
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1402/6/066036
Subject(s) - artificial intelligence , pattern recognition (psychology) , iris recognition , computer science , naive bayes classifier , support vector machine , normalization (sociology) , grayscale , gabor filter , segmentation , computer vision , feature extraction , feature vector , pixel , biometrics , sociology , anthropology
Algorithms developed to identify people with iris image data have been tested in many field and laboratory experiment. This paper analysis some a parameters of iris image used to recognize human. Iris recognition system, which is applied based on segmentation, normalization, encoding, and matching is also describe in this paper. Circle Hough Transform segmentation module used to find the inner and outer boundaries of the iris. The experiment was carried out using CASIA v1 iris database with grayscale images. Shape, intensity, and location information for localizing the pupil or iris and normalizing the iris area a used iris segmentation by unwrapping circular area into a rectangular area. Normalized area will be used to extract the features using Gray Level Co-occurrence Matrix (GLCM) and Gabor filter, the feature compared the recognition accuracy using Support Vector Machines (SVM) and Naive Bayes classifiers. GLCM feature test results achieved 95.24% SVM classification accuracy, whereas using achieved 85.71% Naive Bayes. Gabor feature test results achieved 95.24% SVM classification accuracy, whereas using achieved 95.23% Naive Bayes. The classification process based on GLCM and Gabor features show that the SVM method have to highest recognition accuracy compare to Naive Bayes classifier.