CAJ | 학술논문

针对目前混沌加密数字图像难以抵抗已知/选择明文攻击的主要缺陷，提出了基于级联L-L混沌系统改进的图像加密算法。不同于传统Fridrich置乱和扩散的架构，采用一种新的改进的加密算法架构使置乱-扩散过程相结合，并利用明文MD5值分组异或调制级联混沌系统的参数和初始值，增加算法对明文的敏感性，从而使整个加密算法能有效地抵抗已知/选择明文攻击。实验结果表明：该改进算法有着很好的加密效果和安全性，并在明文敏感性分析方面效果更好，能有效抵抗已知/选择明文攻击。
침대목전혼돈가밀수자도상난이저항이지/선택명문공격적주요결함，제출료기우급련L-L혼돈계통개진적도상가밀산법。불동우전통Fridrich치란화확산적가구，채용일충신적개진적가밀산법가구사치란-확산과정상결합，병이용명문MD5치분조이혹조제급련혼돈계통적삼수화초시치，증가산법대명문적민감성，종이사정개가밀산법능유효지저항이지/선택명문공격。실험결과표명：해개진산법유착흔호적가밀효과화안전성，병재명문민감성분석방면효과경호，능유효저항이지/선택명문공격。
Aiming at the main defect that the present digital image encryption algorithm based on chaotic is difficult to resist the known/choose plaintext attack, this paper proposes an improved algorithm based on cascade chaotic system. Different from traditional Fridrich architecture of confusion and diffusion, an improved encryption algorithm architecture with con-fusion-diffusion process is proposed, and the original image’s MD5 value is XORed to control cascade chaotic system’s parameters and initial values respectively, increasing the encryption algorithm’s sensitivity of original image, as a result the encryption algorithm can effectively resist known/choice plaintext attack. Experimental results show that the improved algorithm has good encryption efficacy and security, and better results in original image’s sensitivity analysis, can effec-tively resist known/choose plaintext attack.