文章基本信息

标题：Novel Approach for Face Recognition Using Fractals
本地全文：下载
作者：Mr. Rakesh S. Shirsath ; Prof. L. B. Randive
期刊名称：International Journal of Electronics, Communication and Soft Computing Science and Engineering
印刷版ISSN：2277-9477
出版年度：2014
卷号：3
期号：3
出版社：IJECSCSE
摘要：Biometrics has been used for personidentification since long time. Till today, researchers,neuroscientists and engineers working on the how humanbrain works for identification of the person. Hence it is avery interesting topic to do the work on it.Every individual has unique fingerprint and it doesnot changes throughout the lifetime. So it is the most widelyused technique. But the face is most natural, non intrusivebiometric and the cost of the system is low as compared withthe other biometric system.Fractals are popular because of their ability tocreate complex images using only several simple codes. Thisis possible by capturing image redundancy and presentingthe image in compressed form using the self similarityfeature. In this paper we present new fractal method forhuman face recognition. Many conventional methods of facerecognition depend solely on appearance and model based.But there is inbuilt degree of self similarity in the image offaces which can be efficiently utilized through representationexploiting self-transformations, known as Iterated FunctionSystem (IFS).Interestingly, virtually all images of natural orman-made objects, show region wise self similarity althoughthey may not be globally self similar. Such objects can berepresented by Partitioned Iterated Function System (PIFS)very compactly.