الفهرس 
Chapter : IntroductionOnly 14 pages are availabe for public view
 |  | from | 161 |  |  |
| | | from | 161 | | |
Abstract
Cryptography is a serious merit of computer and network security. Using biometrics for security purposes is being more common, but utilizing biometrics by means of cryptography is a modern, increasing and promising research area. Thus, we will suggest some algorithms that are more renitent to attacks and security sensibilities are removed progressively. Due to an accepted encryption algorithm must impede all types of cryptanalytic attacks such as that statistical attacks and brute force attacks, differential attacks and related key attacks. In this research, a novel methodology for securing digital images is proposed which utilizes two types of chaotic systems (Logistic map and Tent map). The main feature of the proposed scheme lies on the utilization of biometric secret image to drive an external secret key, which in turn deduces the control parameters of the maps. Also, in this research we propose a new method for image encryption that has fast performance speed and high level security. The design tools of our algorithm are based on a chaos-based feedback cryptographic scheme using the tent map and an external biometric secret key of -bit using hash function (SHA- ); moreover, we introduce a new image encryption approach using Tinkerbell map and extract minutia points of fingerprint as Key by using image processing algorithm. The design tools of our algorithm are based on a chaos-based feedback cryptographic scheme using the Tinkerbell map and an external biometric secret key using Minutiae Extraction of fingerprint as Key.
Encryption algorithm must be further strong against all types of cryptanalytic attacks such statistical attacks and exhaustive search attacks, differential attacks and besides related key attacks. We check the proposed algorithms against statistical attacks, exhaustive search attacks, differential attacks and related key attacks. The results appeared that the cipher is extremely susceptible to changes in the encryption key and the plain image. Therefore, the proposed schemes is robust against the popular attacks. In addition, we compared our results of proposed schemes with the standard cryptography algorithms (AES, RC , and RC ); the results demonstrate that proposed schemes keep the smallest values which explain that the encrypted image has small correlations among neighbor pixels, almost united image histogram which can be seen as nearly random image. So the proposed schemes are resistant to statistical and differential offensives.