Biometrics lack revocability and privacy while cryptography cannot detect the user’s identity. By obtaining cryptographic keys using biometrics, one can achieve the properties such as revocability, assurance about user’s identity, and privacy. In addition, the Multi-biometric systems alleviate a few of the problems observed in unimodal biometric systems. Besides improving matching performance and its can integrate information at various levels.
In this thesis presents Document Signing Using ESSO Algorithm, its aims to introduce a new technique to generate stream cipher key by using a multi –biometric identification system that consists of (sclera and palm) images and using the best coordinates is produced by an enhancement of shark smell optimization algorithm (SSO) based on 3d logistic chaotic map.
The stages of implementation the proposed system include feature extraction from sclera and palm biometric images of same personal using proposed sclera –identification system and palm –identification system, each system has different preprocessing techniques to preparing images to exacted features.
In addition , the scale-invariant feature transform (SIFT) algorithm in multibiometric identification system extract features from the sclera and palm biometric images and to get fast access to the important regions inside the sclera and palm biometric images. These features are used to find the optimal solution by using enhancement shark smell optimization algorithms(SSO) based on chaotic function.
The enhancements shark smell optimization algorithm (SSO)consists from three steps: 3d logistic map to generate set of random numbers to seed random parameters (R1,R2,R3) in SSO algorithm this step aims to enhance the performance of SSO , the features that extracted from sclera and palm biometric images dropping on secret image to connected between sclera feature that represent (shark position ) and palm feature that represents ( fish position ) this step aims to find set of optimal solution .
To generate stream cipher key, the proposing system presents new technique to generating variable and unpredictable stream key based on convert to binary the values of all optimal solution (fitness value, objective value, solution point coordinate (x,y)) .
The final stage in the proposed system is the document signature by using the MD5, the proposed system has the ability to generate a unique digital signature for each user.
The proposed system using the stream key in document signature to protect personal information in a completely safe manner.
The implementation the proposed system and results of Random Number Generation Tests (NIST) National Institute of Standards and Technology shown the proposed system has ability to generating stream key for multiply users that has several proprieties likes: unique, unpredictable, strong, and various length.