期刊名称：International Journal of Advanced Research In Computer Science and Software Engineering
印刷版ISSN：2277-6451
电子版ISSN：2277-128X
出版年度：2013
卷号：3
期号：4
出版社：S.S. Mishra
摘要：The o bjective of this paper is to present an accurate technique to identify two lookalike faces based on the concepts of biometrics and the Median Filter Algorithm. This paper gives some useful insight on the application of median filter and biometrics, where biometrics refers to methods for uniquely recognizing humans based upon one or more intrinsic physical or behavioral traits. It is also used to identify individuals in groups that are under surveillance. Biometric recognition based on human characteristics for personal identification has attracted great attention. The performance of biometric systems highly depends on the distinctive information in the biometrics. However, lookalike faces having the closest genetics-based relationship are expected to have maximum similarity between their biometrics. Identifying such lookalike faces is a challenging problem for most automatic biometric systems. Physiological biometrics has been used to operate on still images of the face. The challenge is to get rid of high degree of noise in the images before any further processing is done. The median filter is used for this purpose in this project which is a non-linear digital filtering technique. The project has been successfully implemented with an efficiency of about 85%.
关键词：Lookalike Face Identification; Biometric; Median Filter
