MultiModal Identification System in Monozygotic Twins

International Journal of Image Processing

Dinakar das CN¹  Nisha George¹  S.Peumal Sankar¹ 

[1] $$

关键词: Fisher Faces;    Principal Component Analysis;    Local Binary Pattern;    Receiver Operating Characteristics Curve.;   

DOI  :  

来源: Computer Science Journals

【 摘 要 】

With the increase in the number of twin births in recent decades, there is a need to developalternate approaches that can secure the biometric system. In this paper an effective fusionscheme is presented that combines information presented by multiple domain experts based onthe rank-level fusion integration method. The developed multimodal biometric system possessesa number of unique qualities, starting from utilizing Fisher’s Linear Discriminant methods for facematching, Principal Component Analysis for fingerprint matching and Local binary patternfeatures for iris matching and fused the information for effective recognition and authenticationThe importance of considering these boundary conditions, such as twins, where the possibility oferrors is maximum will lead us to design a more reliable and robust security system.Theproposed approach is tested on a real database consisting of 50 pair of identical twin images andshows promising results compared to other techniques. The Receiver Operating Characteristicsalso shows that the proposed method is superior compared to other techniques under study.

【 授权许可】

Unknown