文章基本信息

标题：Identification of Fingerprint Using Discrete Wavelet Transform in Conjunction with Support Vector Machine
本地全文：下载
作者：Shahid Akbar ; Ashfaq Ahmad ; Maqsood Hayat 等
期刊名称：International Journal of Computer Science Issues
印刷版ISSN：1694-0784
电子版ISSN：1694-0814
出版年度：2014
卷号：11
期号：5
出版社：IJCSI Press
摘要：Fingerprint recognition is mostly used in biometric and security system that are practically applicable in different fields of national defense organization for various safety measures. Looking at the importance of biometric system, a lot of efforts have been carried out for recognition of fingerprint, but still there exist some issues, which demand for more attention and exploration. In this regards, we attempt to develop a vigorous and reliable biometric system for recognition of fingerprint. The proposed computational model is designed using three discrete features extraction methods such as Discrete Wavelet transform, Principal Component Analysis and Discrete Cosine Transform. Two diverse nature of classification hypothesis are utilized namely: Support Vector Machine (SVM) and K-nearest neighbor. Three different benchmark fingerprint datasets and 10-folds cross validation are applied to evaluate the performance of the proposed model. The empirical results reveal that SVM has achieved outstanding performance using all the three benchmark datasets. It is ascertained that the predicted model may be useful and high throughput tool for academia and security related areas.
关键词：DWT; PCA; DCT; KNN; SVM; 10;folds