期刊名称：Signal & Image Processing : An International Journal (SIPIJ)
印刷版ISSN：2229-3922
电子版ISSN：0976-710X
出版年度：2012
卷号：3
期号：3
页码：103
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：Recognition of newborn after birth is a critical issue for hospitals, maternity word and other places wheremultiple birthstake place. Switching and abduction of newborn is a global challenge and research done tomeet this challenge is minimal. Most of the biometric systems developed are for adults and very few of themaddress the issue of newborn identification.The ear of newborn is a perfect source of data for passiveidentification of newborn as they are the highly non cooperative users of biometrics. The four importantcharacteristics of ear biometrics: universality, uniqueness, permanence and collectability make it a verypotential biometric trait for the identification of newborn. Further the use of soft-biometric data likegender, blood group, height and weight along with ear enhances the accuracy for identification ofnewborn. The objective of this paper is to demonstrate the concept of fusing earand soft-biometrics forrecognition of newborn. The main contributions of the research are (a) Design and implementation forfusion of ear and soft biometric for recognition of 210 newborn. (b)Preparation of ear and soft-biometricdatabase of newborn. Fusion of ear and soft-biometrics results in an improvement of approximately 5.59%over the primary biometric system i.e. ear.
