Optimizasyon test fonksiyonlarının hibritleştirilmesi ile tasarlanan iki boyutlu hiperkaotik harita kullanılarak çapraz kanal renkli görüntü şifreleme algoritmasının geliştirilmesi
Tez Türü: Doktora
Tezin Yürütüldüğü Kurum: Mersin Üniversitesi, Fen Bilimleri Enstitüsü, Bilgisayar Mühendisliği, Türkiye
Tez Danışmanı: Uğur Erkan; Zeki Yetgin
Tezin Onay Tarihi: 2024
Tezin Dili: Türkçe
Özet:
Chaos-based Image Encryption (IE) has emerged as a superior alternative by exploiting the inherent unpredictability and sensitivity to initial conditions inherent in chaotic systems. The chaotic maps that underpin this approach improve security by introducing complex and seemingly random changes to image data. To be effective, an IE chaotic map must exhibit key properties such as ergodicity, high sensitivity to initial conditions, and scrambling, which ensures that even the smallest change in the input results in a very different output, thus providing robust protection against cyber-attacks. The security of chaos-based IE algorithms inherently depends on the permutation and diffusion strategy of the algorithm and the dynamic performance of chaotic maps. However, the existing studies suffer from low disordering capability, poor dynamic performance, and lack of sensitivity of the chaotic maps. In this thesis, a new hyperchaotic map with better chaotic properties is designed to overcome the limitations of existing chaotic maps and a secure IE algorithm is developed. By exploiting the chaotic properties of optimization test functions such as high complexity and fluctuation, a hyperchaotic map is designed by hybridizing two-dimensional (2D) Rastrigin and Griewank (2D-RG) functions. The 2D-RG map has been measured with the methods effectively used in dynamical analysis and the results have been compared with the maps in the literature. In addition, a cross permutation and bidirectional sequential diffusion Cross-Channel Color IE (CCC-IE) algorithm with high confusing capability using the 2D-RG map is developed. While encrypting, the 2D-RG map-based CCC-IE algorithm scrambles the pixels of the image diagonally along the colored RGB (Red-Green-Blue) channels and thus alternately changes the pixel values in two directions to increase the algorithm's robustness against cyber threats. To demonstrate the superiority of the proposed cryptosystem, recent hyperchaotic maps and IE algorithms in the literature are elaborately analyzed. In this context, the dynamic performance of the 2D-RG map and the reliability of the KAR-GŞ algorithm are analyzed separately and a detailed comparison is made with recent methods in the literature. The results show that the KAR-GŞ algorithm is a more secure algorithm against cyber-attacks in both visual and numerical analysis thanks to the superior chaotic properties of the 2D-RG map.