19-08-2014, 10:50 AM
LITERATURE SURVEY
LITERATURE SURVEY.doc (Size: 566.5 KB / Downloads: 10)
.INTRODUCTION
1.1. GENERAL
PALM-PRINT recognition has now been a topic of research for over 10 years. Like other biometrics, palm prints demonstrate the properties required for personal authentication: universality, uniqueness, permanence, collectability, and acceptability. Furthermore, palm prints have some advantages over other biometrics. Palm prints are larger than fingerprints and therefore more robust to scars and dirt. Palm-print images are cheaper to collect and more acceptable than iris. Palm prints can distinguish between individuals more accurately than faceand can also identify monozygotic twins. Traditionally, palm-print recognition has made use of either high- or low-resolution 2-D palm-print images. High-resolution images are suitable for forensic applications, while low-resolution images are suitable for civil and commercial applications. Most current research uses low-resolution palm-print recognition and is either texture based or line based. The texture-based methods include PalmCode, Competitive Code, and Ordinal Code. These methods use a group of filters to enhance and extract the phase or directional features which can represent the texture of the palm print. Linebased methods use line or edge detectors to explicitly extract line information from the palm print that is then used for matching. The representative methods include derivative-of- Gaussian-based line extraction and modified-finite-Radontransform- based line extraction.
1.2 Existing method
• Authentication system with single biometric modality
• Finger printing is that much not flexible because we can make duplicates of fingers and bluff people. It is not that much efficient.
• RSA encryption and pixel difference expansion based embedding
1.3 Drawbacks
• Uni modal biometric system may not fulfill the requirements of demanding
Applications interms of properties such as performance, acceptability
• More computational complexity
• High distortion and algorithm complexity
1.4 Proposed method
Person authentication system based on multimodal biometrics and steganographic with cryptography technology, it involves
• Iris with palmprint recognition
• Chaos encryption
• Data concealment using adaptive lsb algorithm
1.5 Advantages
• Low distortion due to data concealment
• Better data protection and less complexity
• This system gives more security compared to unimodal system because of two biometric features
Irispattern Extraction Using Bitplanes And Standard Deviations:
In describe(1) Iris recognition has been shown to be very accurate for human identification. In this paper, we develop a technique for iris pattern extraction utilizing the least significant bit-plane: the least significant bit of every pixel in the image. Through binary morphology applied to the bit-plane, the pupillay boundary of the iris is determined. The limbic boundary is identified by evaluating thestandard deviation of the image intensity along the verticaland horizontal axes. Because our extraction approach restricts localization techniques to evaluating only bitplanes and standard deviations, iris pattern extraction is not dependent on circular edge detection. This allows for an expanded functionality of iris identification technology by no longer requiring a frontal view, which leads to the potential for off-angle iris recognition technology. Initial results show that it is possible to fit a close elliptical approximation to an iris pattern by using only bit-planesand standard deviations for iris localization.
3 Embedding:
In the data embedding phase, some parameters are embedded into a small number of encrypted pixels, and the LSB of the other encrypted pixels are compressed to create a space for accommodating the additional data and the original data at the positions occupied by the parameters. The detailed procedure is as follows According to a data-hiding key, the data-hider pseudo-randomly. In encryption phase, the exclusive-or results of the original bits and pseudo-random bits are calculated. selects Np encrypted pixels that will be used to carry the parameters for data hiding. Here, Np is a small positive integer, for example, Np=20. The other (N-Np) encrypted pixels are pseudo-randomly permuted and divided into a number of groups, each of which contains L pixels. The permutation way is also determined by the data-hiding key. For each pixel-group, collect the M least significant bits of the L pixels, and denote them as B (k,1) , B (k,2) …… B(k,M*L) where k is a group index within [1,(N-Np)/L] and M is a positive integer less than 5. The data-hider also generates a matrix G sized (M*L – S) * M*L, which is composed of two parts. The left part is the identity matrix and the right part is pseudo-random binary matrix derived from the data-hiding key. For each group , which is product with the G matrix to form a matrix of size (M * L-S). Which has a sparse bits of size S, in which the data is embedded and arrange the pixels into the original form and repermutated to form a original image.
Feature Extraction :
Features are the attributes or values extracted to get the unique characteristics from the image. Features from the iris image and palmprint image are extracted using Wavelet decomposition process. In the wavelet decomposition the image is decomposed into four coefficient i.e., horizontal, diagonal, vertical and approximation. The approximation coefficients are further decomposed into four coefficients. The sequences of steps are repeated for five levels and the last level coefficients are combined to form a vector.
3.9 Decryption:
When having an encrypted image containing embedded data, a receiver firstly generates ri,j,k according to the encryption key, and calculates the exclusive-or of the received data and ri,j,k to decrypt the image. We denote the decrypted bits as b1i,j,k . Clearly, the original five most significant bits (MSB) are retrieved correctly. For a certain pixel, if the embedded bit in the block including the pixel is zero and the pixel belongs to S1, or the embedded bit is 1 and the pixel belongs to S0 , the data-hiding does not affect any encrypted bits of the pixel. So, the three decrypted LSB must be same as the original LSB, implying that the decrypted gray value of the pixel is correct. On the other hand, if the embedded bit in the pixel’s block is 0 and the pixel belongs to S0, or the embedded bit is 1 and the pixel belongs to S1 , the decrypted LSB
CONCLUSION
The proposal of this work is a three global features for 3-D palmprint. These cannot be extracted from 2-D palm prints and are not correlated with local features such as line and texture features. To make these global features efficient for use in coarse classification, we treat them as a multidimensional vector and use OLDA to map it to a lower dimensional space. We then improve the efficiency of 3-D palmprint recognition using two proposed approaches, coarse-level matching and RSVM, both of which significantly reduce the penetration rate during retrieval. Our recognition experiments using an established 3-D palm-print database of 8000 samples show that the global features improve palm-print classification which greatly reduces search times.