JPEG彩色图像自适应加密算法  

Self-adaptive Encryption for JPEG Color Images

作　　者：文昌辞[1,2] 王沁[1] 黄付敏[3] 刘向宏 陈新中 

机构地区：[1]北京科技大学计算机科学与技术系,北京100083 [2]空军驻京昌地区军事代表室,北京100041 [3]中国医学科学院北京协和医学院,北京100730 [4]中国电子科学研究院院办公室,北京100041

出　　处：《计算机辅助设计与图形学学报》2012年第4期500-505,共6页Journal of Computer-Aided Design & Computer Graphics

摘　　要：基于复合混沌和有限整数域上的仿射变换,提出一种结合JPEG图像压缩编码的加密算法.首先在空域对R,G,B分量以8×8大小块为基本单元统一进行位置置乱,打乱R,G,B分量之间的组合关系,再进行正常的JPEG压缩编码;在量化DCT系数之后,对亮度、色度分量中的DC系数分别进行置乱,置乱系数位置的同时根据坐标混合它的值,然后扰动复合混沌系统以自适应地代换DC系数.该算法符合模块化设计,密钥空间大、安全性高.实验结果表明,文中算法视觉效果好、敏感性强,密文与直接压缩的图像大小相当.Based on the compound chaos and affine transformation in finite integer domain,an encryption algorithm is proposed to embed into the compression of JPEG image.Firstly,the blocks of 8×8 pixels of R,G and B components are scrambled globally in space domain to disarrange the combination relation among three components before the precedence of normal JPEG compression.Secondly,the frequency domain＇s DC coefficients of brightness and color components are scrambled respectively after quantization,and their values are mixed according to the corresponding coordination.Lastly,the compound chaos system is perturbed to substitute for the DC coefficients self-adaptively.Our algorithm is designed modularly with huge key space and high security.The experiments showed that the algorithm produce results with high visual quality with strong sensitivity,and could generate the ciphertext with almost the same size as that of the directly compressed image.

关 键 词：加密 JPEG 混沌 置乱 代换 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]