基於Margolus 細胞自動機結合通道隨機排列之可逆圖像加密方法

Abstract

細胞自動機是一種離散模型，已經成功地應用在許多物理系統及科 學問題中，例如數學、生物學都有其相關研究。它的三個重要特徵： 平行計算、變化局部性及規則一致性使得其非常適合應用在圖像加密 上，大多數現代圖像加密方法依賴於靜態數學原理來模糊數據，相比 之下細胞自動機不是基於公式來演化，它依賴於先前的鄰域狀態進行 下一個狀態的計算，具有混亂及不可預測性。本論文提出了一種基於 Margolus 細胞自動機之可逆圖像加密方法，Margolus 細胞自動機易於 找到可逆規則的特性使得密鑰可簡單得到，方法結合了R、G、B 顏色 通道之隨機排列增加整體的加密複雜度。本論文針對加密後之圖像做 了多種實驗分析，結果顯示所提出的方法具有高度的安全性可抵擋差 異攻擊及暴力攻擊，並且加密圖像可以零失真還原，平行化計算也使 得方法與序列執行相比具有更高的效率。

Cellular automata represent a discrete model which has been successfully applied to many physical systems and scientific problems in mathematics and biology. Its three important features, namely parallel computing, changing locality and regularity, make it ideal for image encryption. Most modern image encryption methods rely on static mathematical principles to obfuscate data, whereas cellular automatons do not rely on the evolution of formulas, instead, they use previous neighborhood states for the next state calculation, exhibiting randomness and unpredictability. A new reversible color image encryption method based on Margolus cellular automata is proposed. Margolus cellular automata make it easy to find reversible rules. The method combines the random shuffle of R, G, B color channels to increase the overall encryption complexity. By conducting experiment analysis on a wide variety of encrypted images, our results show that the proposed method has a high degree of security to withstand differences in differential attacks and violent attacks. Furthermore, the encrypted image can be decrypted without data loss, and parallel calculation also makes the method efficient.