期刊名称：International Journal of Electrical and Computer Engineering
电子版ISSN：2088-8708
出版年度：2013
卷号：3
期号：1
页码：30-41
DOI：10.11591/ijece.v3i1.1760
语种：English
出版社：Institute of Advanced Engineering and Science (IAES)
摘要：The pattern of the dorsal hand vein is a new sustainable, unique and resistant biometric approach which has been recently attracted the attention of researchers. Due to the extensive studies on biometrics like fingerprint, face, ears, palms vessels, iris and human gait, dorsal hand vein is more preferable than other types of biometrics. The fingerprint can just be obtained by touching a surface by an individual. So it is possible to steal or counterfeit the patterns as by the first touch, the sensor might become dirty and so reducing the image quality may cause poor results. On the other hand, the pattern of the vessels of back of the hand is fixed and unique with repeatable biometric features. Also, due to noises in the imaging patterns, and impossibility of further reducing because of the non-complexity of the models, no certain recognition rate has been obtained yet, and proof of correctness of identification is required. Therefore, in this paper, first, the images of blood vessels on back of the hands of people is analysed, and after pre-processing of images and feature extraction (in the intersection between the vessels) we began to identify people using firefly clustering algorithms. This identification is done based on the distance patterns between crossing vessels and their matching place. The identification will be done based on the classification of each part of NCUT data set and it consisting of 2040 dorsal hand vein images. High speed in patterns recognition and less computation are the advantages of this method. The recognition rate of this method is more accurate and the error is less than one percent. At the end the correctness percentage of this method (CLU-D-F-A) for identification is compared with other various algorithms, and the superiority of the proposed method is proved.
关键词：Computer Science;Biometric, Feature Extraction, Patterns of Dorsal hand vein, Clustering, Firefly Algorithm.
