data. In addition to that identication also done to provide the correct service to the person.

implemented and simulated in MATLAB. The parameters correlation, computation time,

Unied averaged changed intensity(UACI), Number of changing pixel rate(NCPR)and

Entropy are calculated and the results show the eectiveness of encrypted fusion based on

DWT. Entropy has been increased by 67%. NCPR of 99.6239 and UACI of 3.3004.

Fusion, Face biometric, Iris biometric, DWT, Encryption.

which two images are fused together to obtain a single image. Images with dierent focused

regions, images from dierent modalities or images taken in dierent times have been

by Stationary Wavelet Transform (SWT), Discrete Wavelet Transform (DWT), Lifting

Wavelet Transform (LWT), Redundancy Discrete Wavelet Transform (RDWT), Dual-Tree

Complex Wavelet Transform (DTDWT). These methods have unique drawbacks but all

A multimodal biometric system is developed using ngerprint and iris biometric in Rajbhoj

and Mane (2015). The system combines ngerprint and iris at feature level. Single feature

vector is obtained by fusing ngerprint and iris image and unique textural pattern from

fused image is obtained by ecient wavelet transform. Matching is carried using Hamming

they used correlation coecient as the similarity measure. A multimodal biometric system

is developed by combining face and ngerprint biometric by score level fusion. According

to \cite{19} face recognition is done using PCA and ngerprint recognition is done using

minutiae matching and Gabor ltering.

print and ngerprint in feature level. Palm print and nger print images were fused using

wavelet based image fusion techniques with min-min approximation. Features were extracted

using Discrete Cosine Transform (DCT) and feature reduction is done. In their work also

independent databases are used for palm print and ngerprint and they are combined

by assigning a ngerprint image to each palm print image. Their shows that multi modal

Our project gives the following signicant works

The remaining sections are organized as follows: section 2 provides the proposed methods

The proposed DWT consists of four steps: Enhancement, decomposition, fusion and

encryption. The block diagram of the proposed DWT-based face iris fusion is shown in

and diagonal details have been found out for two levels for both images. The individual sub

The Haar wavelet illustrates the desirable properties of wavelets in general. It can

. DWT demonstrates the localization: the rst row (1,1,1,1) term gives

domain; and (1,-1,0,0) places it at the left of the left side. The explanation of DWT is

low pass lter with impulse response g (n), Hence the convolution operation between input

There is a unique set of expansion coecients in every representable function of Fourier

The performance of biometric system is improved by choosing correct fusion. In our

approach feature level fusion is used. The feature vectors at second level of DWT are

combined to produce the fused image. For verication of individual persons biometric,

dierent images of any biometric can be fused. But authentication purpose, two dierent

initial key is set as “AAAAAAAAAAAA” for this algorithm. It can be expanded for next

rounds. The key expansion diagram is shown in Figure 2. The confusion and diusion are

If j=2 second operation is performed. If j=3 third operation is done. The nal encryption

operation is performed as the concatenation result of fth round.

Encryption algorithm to do the operations in a simple manner without the need of S-box

Encryption. Cyber Journals: Multidisciplinary Journals in Science and Technology,